Permalink
Fetching contributors…
Cannot retrieve contributors at this time
6278 lines (5392 sloc) 106 KB
% !Mode:: "TeX:UTF-8"
\newglossaryentry{DL}
{
name=深度学习,
description={deep learning},
sort={deep learning},
}
\newglossaryentry{knowledge_base}
{
name=知识库,
description={knowledge base},
sort={knowledge base},
}
\newglossaryentry{ML}
{
name=机器学习,
description={machine learning},
sort={machine learning},
}
\newglossaryentry{ML_model}
{
name=机器学习模型,
description={machine learning model},
sort={machine learning model},
}
\newglossaryentry{ebv}
{
name=基本单位向量,
description={elementary basis vectors},
sort={elementary basis vectors},
}
\newglossaryentry{logistic_regression}
{
name=逻辑回归,
description={logistic regression},
sort={logistic regression},
}
\newglossaryentry{regression}
{
name=回归,
description={regression},
sort={regression},
}
\newglossaryentry{AI}
{
name=人工智能,
description={artificial intelligence},
sort={artificial intelligence},
symbol={AI}
}
\newglossaryentry{naive_bayes}
{
name=朴素贝叶斯,
description={naive Bayes},
sort={naive Bayes},
}
\newglossaryentry{representation}
{
name=表示,
description={representation},
sort={representation},
}
\newglossaryentry{representation_learning}
{
name=表示学习,
description={representation learning},
sort={representation learning},
}
\newglossaryentry{AE}
{
name=自编码器,
description={autoencoder},
sort={autoencoder},
}
\newglossaryentry{encoder}
{
name=编码器,
description={encoder},
sort={encoder},
}
\newglossaryentry{decoder}
{
name=解码器,
description={decoder},
sort={decoder},
}
\newglossaryentry{MLP}
{
name=多层感知机,
description={multilayer perceptron},
sort={multilayer perceptron},
symbol={MLP}
}
\newglossaryentry{cybernetics}
{
name=控制论,
description={cybernetics},
sort={cybernetics},
}
\newglossaryentry{connectionism}
{
name=联结主义,
description={connectionism},
sort={connectionism},
}
\newglossaryentry{ANN}
{
name=人工神经网络,
description={artificial neural network},
sort={artificial neural network},
symbol={ANN}
}
\newglossaryentry{NN}
{
name=神经网络,
description={neural network},
sort={neural network},
}
\newglossaryentry{SGD}
{
name=随机梯度下降,
description={stochastic gradient descent},
sort={stochastic gradient descent},
symbol={SGD}
}
\newglossaryentry{linear_model}
{
name=线性模型,
description={linear model},
sort={linear model},
}
\newglossaryentry{mode}
{
name=峰值,
description={mode},
sort={mode},
}
\newglossaryentry{unimodal}
{
name=单峰值,
description={unimodal},
sort={unimodal},
}
\newglossaryentry{modality}
{
name=模态,
description={modality},
sort={modality},
}
\newglossaryentry{multimodal}
{
name=多峰值,
description={multimodal},
sort={multimodal},
}
\newglossaryentry{linear_regression}
{
name=线性回归,
description={linear regression},
sort={linear regression},
}
\newglossaryentry{ReLU}
{
name=整流线性单元,
description={rectified linear unit},
sort={rectified linear unit},
symbol={ReLU}
}
\newglossaryentry{distributed_representation}
{
name=分布式表示,
description={distributed representation},
sort={distributed representation},
}
\newglossaryentry{nondistributed_representation}
{
name=非分布式表示,
description={nondistributed representation},
sort={nondistributed representation},
}
\newglossaryentry{nondistributed}
{
name=非分布式,
description={nondistributed},
sort={nondistributed},
}
\newglossaryentry{hidden_unit}
{
name=隐藏单元,
description={hidden unit},
sort={hidden unit},
}
\newglossaryentry{LSTM}
{
name=长短期记忆,
description={long short-term memory},
sort={long short-term memory},
symbol={LSTM}
}
\newglossaryentry{DBN}
{
name=深度信念网络,
description={deep belief network},
sort={deep belief network},
symbol={DBN}
}
\newglossaryentry{RNN}
{
name=循环神经网络,
description={recurrent neural network},
sort={recurrent neural network},
symbol={RNN}
}
\newglossaryentry{recurrence}
{
name=循环,
description={recurrence},
sort={recurrence},
}
\newglossaryentry{RL}
{
name=强化学习,
description={reinforcement learning},
sort={reinforcement learning},
}
\newglossaryentry{inference}
{
name=推断,
description={inference},
sort={inference},
}
\newglossaryentry{overflow}
{
name=上溢,
description={overflow},
sort={overflow},
}
\newglossaryentry{underflow}
{
name=下溢,
description={underflow},
sort={underflow},
}
\newglossaryentry{softmax}
{
name=softmax函数,
description={softmax function},
sort={softmax function},
}
\newglossaryentry{softmax_chap15}
{
name=softmax,
description={softmax},
sort={softmax},
}
\newglossaryentry{underestimation}
{
name=欠估计,
description={underestimation},
sort={underestimation},
}
\newglossaryentry{overestimation}
{
name=过估计,
description={overestimation},
sort={overestimation},
}
\newglossaryentry{softmax_unit}
{
name=softmax单元,
description={softmax unit},
sort={softmax unit},
}
\newglossaryentry{softmax_chap11}
{
name=softmax,
description={softmax},
sort={softmax},
}
\newglossaryentry{multinoulli}
{
name=Multinoulli分布,
description={multinoulli distribution},
sort={multinoulli distribution},
}
\newglossaryentry{poor_conditioning}
{
name=病态条件,
description={poor conditioning},
sort={poor conditioning},
}
\newglossaryentry{objective_function}
{
name=目标函数,
description={objective function},
sort={objective function},
}
\newglossaryentry{objective}
{
name=目标,
description={objective},
sort={objective},
}
\newglossaryentry{criterion}
{
name=准则,
description={criterion},
sort={criterion},
}
\newglossaryentry{cost_function}
{
name=代价函数,
description={cost function},
sort={cost function},
}
\newglossaryentry{cost}
{
name=代价,
description={cost},
sort={cost},
}
\newglossaryentry{loss_function}
{
name=损失函数,
description={loss function},
sort={loss function},
}
\newglossaryentry{prcurve}
{
name=PR曲线,
description={PR curve},
sort={PR curve},
}
\newglossaryentry{fscore}
{
name=F分数,
description={F-score},
sort={F-score},
}
\newglossaryentry{loss}
{
name=损失,
description={loss},
sort={loss},
}
\newglossaryentry{error_function}
{
name=误差函数,
description={error function},
sort={error function},
}
\newglossaryentry{GD}
{
name=梯度下降,
description={gradient descent},
sort={gradient descent},
}
\newglossaryentry{local_descent}
{
name=局部下降,
description={local descent},
sort={local descent},
}
\newglossaryentry{steepest}
{
name=最陡下降,
description={steepest descent},
sort={steepest descent},
}
\newglossaryentry{GA}
{
name=梯度上升,
description={gradient ascent},
sort={gradient ascent},
}
\newglossaryentry{derivative}
{
name=导数,
description={derivative},
sort={derivative},
}
\newglossaryentry{critical_points}
{
name=临界点,
description={critical point},
sort={critical point},
}
\newglossaryentry{stationary_point}
{
name=驻点,
description={stationary point},
sort={stationary point},
}
\newglossaryentry{local_minimum}
{
name=局部极小点,
description={local minimum},
sort={local minimum},
}
\newglossaryentry{minimum}
{
name=极小点,
description={minimum},
sort={minimum},
}
\newglossaryentry{local_minima}
{
name=局部极小值,
description={local minima},
sort={local minima},
}
\newglossaryentry{minima}
{
name=极小值,
description={minima},
sort={minima},
}
\newglossaryentry{global_minima}
{
name=全局极小值,
description={global minima},
sort={global minima},
}
\newglossaryentry{local_maxima}
{
name=局部极大值,
description={local maxima},
sort={local maxima},
}
\newglossaryentry{maxima}
{
name=极大值,
description={maxima},
sort={maxima},
}
\newglossaryentry{local_maximum}
{
name=局部极大点,
description={local maximum},
sort={local maximum},
}
\newglossaryentry{saddle_points}
{
name=鞍点,
description={saddle point},
sort={saddle point},
}
\newglossaryentry{global_minimum}
{
name=全局最小点,
description={global minimum},
sort={global minimum},
}
\newglossaryentry{partial_derivatives}
{
name=偏导数,
description={partial derivative},
sort={partial derivative},
}
\newglossaryentry{gradient}
{
name=梯度,
description={gradient},
sort={gradient},
}
\newglossaryentry{identifiable}
{
name=可辨认的,
description={identifiable},
sort={identifiable},
}
\newglossaryentry{directional_derivative}
{
name=方向导数,
description={directional derivative},
sort={directional derivative},
}
\newglossaryentry{line_search}
{
name=线搜索,
description={line search},
sort={line search},
}
\newglossaryentry{example}
{
name=样本,
description={example},
sort={example},
}
\newglossaryentry{hill_climbing}
{
name=爬山,
description={hill climbing},
sort={hill climbing},
}
\newglossaryentry{ill_conditioning}
{
name=病态,
description={ill conditioning},
sort={ill conditioning},
}
\newglossaryentry{jacobian}
{
name=Jacobian,
description={Jacobian},
sort={Jacobian},
}
\newglossaryentry{hessian}
{
name=Hessian,
description={Hessian},
sort={Hessian},
}
\newglossaryentry{second_derivative}
{
name=二阶导数,
description={second derivative},
sort={second derivative},
}
\newglossaryentry{curvature}
{
name=曲率,
description={curvature},
sort={curvature},
}
\newglossaryentry{taylor}
{
name=泰勒,
description={taylor},
sort={taylor},
}
\newglossaryentry{second_derivative_test}
{
name=二阶导数测试,
description={second derivative test},
sort={second derivative test},
}
\newglossaryentry{newton_method}
{
name=牛顿法,
description={Newton's method},
sort={Newton's method},
}
\newglossaryentry{second_order_method}
{
name=二阶方法,
description={second-order method},
sort={second-order method},
}
\newglossaryentry{first_order_method}
{
name=一阶方法,
description={first-order method},
sort={first-order method},
}
\newglossaryentry{lipschitz}
{
name=Lipschitz,
description={Lipschitz},
sort={Lipschitz},
}
\newglossaryentry{lipschitz_continuous}
{
name=Lipschitz连续,
description={Lipschitz continuous},
sort={Lipschitz continuous},
}
\newglossaryentry{lipschitz_constant}
{
name=Lipschitz常数,
description={Lipschitz constant},
sort={Lipschitz constant},
}
\newglossaryentry{convex_optimization}
{
name=凸优化,
description={Convex optimization},
sort={Convex optimization},
}
\newglossaryentry{nonconvex}
{
name=非凸,
description={nonconvex},
sort={nonconvex},
}
\newglossaryentry{nume_optimization}
{
name=数值优化,
description={numerical optimization},
sort={numerical optimization},
}
\newglossaryentry{constrained_optimization}
{
name=约束优化,
description={constrained optimization},
sort={constrained optimization},
}
\newglossaryentry{feasible}
{
name=可行,
description={feasible},
sort={feasible},
}
\newglossaryentry{KKT}
{
name=Karush–Kuhn–Tucker,
description={Karush–Kuhn–Tucker},
sort={Karush–Kuhn–Tucker},
symbol={KKT}
}
\newglossaryentry{generalized_lagrangian}
{
name=广义Lagrangian,
description={generalized Lagrangian},
sort={generalized Lagrangian},
}
\newglossaryentry{generalized_lagrange_function}
{
name=广义Lagrange函数,
description={generalized Lagrange function},
sort={generalized Lagrange function},
}
\newglossaryentry{equality_constraints}
{
name=等式约束,
description={equality constraint},
sort={equality constraint},
}
\newglossaryentry{inequality_constraints}
{
name=不等式约束,
description={inequality constraint},
sort={inequality constraint},
}
\newglossaryentry{regularization}
{
name=正则化,
description={regularization},
sort={regularization},
}
\newglossaryentry{regularizer}
{
name=正则化项,
description={regularizer},
sort={regularizer},
}
\newglossaryentry{regularize}
{
name=正则化,
description={regularize},
sort={regularize},
}
\newglossaryentry{generalization}
{
name=泛化,
description={generalization},
sort={generalization},
}
\newglossaryentry{generalize}
{
name=泛化,
description={generalize},
sort={generalize},
}
\newglossaryentry{underfitting}
{
name=欠拟合,
description={underfitting},
sort={underfitting},
}
\newglossaryentry{overfitting}
{
name=过拟合,
description={overfitting},
sort={overfitting},
}
\newglossaryentry{bias_sta}
{
name=偏差,
description={bias in statistics},
sort={bias in statistics},
}
\newglossaryentry{BIAS}
{
name=偏差,
description={biass},
sort={biass},
}
\newglossaryentry{bias_aff}
{
name=偏置,
description={bias in affine function},
sort={bias in affine function},
}
\newglossaryentry{variance}
{
name=方差,
description={variance},
sort={variance},
}
\newglossaryentry{ensemble}
{
name=集成,
description={ensemble},
sort={ensemble},
}
\newglossaryentry{estimator}
{
name=估计,
description={estimator},
sort={estimator},
}
\newglossaryentry{weight_decay}
{
name=权重衰减,
description={weight decay},
sort={weight decay},
}
\newglossaryentry{ridge_regression}
{
name=岭回归,
description={ridge regression},
sort={ridge regression},
}
\newglossaryentry{tikhonov_regularization}
{
name=Tikhonov正则,
description={Tikhonov regularization},
sort={Tikhonov regularization},
}
\newglossaryentry{covariance}
{
name=协方差,
description={covariance},
sort={covariance},
}
\newglossaryentry{sparse}
{
name=稀疏,
description={sparse},
sort={sparse},
}
\newglossaryentry{feature_selection}
{
name=特征选择,
description={feature selection},
sort={feature selection},
}
\newglossaryentry{feature_extractor}
{
name=特征提取器,
description={feature extractor},
sort={feature extractor},
}
\newglossaryentry{MAP}
{
name=最大后验,
description={Maximum A Posteriori},
sort={Maximum A Posteriori},
symbol={MAP}
}
\newglossaryentry{pooling}
{
name=池化,
description={pooling},
sort={pooling},
}
\newglossaryentry{dropout}
{
name=Dropout,
description={Dropout},
sort={dropout},
}
\newglossaryentry{monte_carlo}
{
name=蒙特卡罗,
description={Monte Carlo},
sort={Monte Carlo},
}
\newglossaryentry{early_stopping}
{
name=提前终止,
description={early stopping},
sort={early stopping},
}
\newglossaryentry{CNN}
{
name=卷积神经网络,
description={convolutional neural network},
sort={convolutional neural network},
symbol={CNN}
}
\newglossaryentry{mcmc}
{
name=马尔可夫链蒙特卡罗,
description={Markov Chain Monte Carlo},
symbol={MCMC},
sort={Markov Chain Monte Carlo},
}
\newglossaryentry{tempering_transition}
{
name=回火转移,
description={tempered transition},
sort={tempered transition},
}
\newglossaryentry{markov_chain}
{
name=马尔可夫链,
description={Markov Chain},
sort={Markov Chain},
}
\newglossaryentry{harris_chain}
{
name=哈里斯链,
description={Harris Chain},
sort={Harris Chain},
}
\newglossaryentry{minibatch}
{
name=小批量,
description={minibatch},
sort={minibatch},
}
\newglossaryentry{importance_sampling}
{
name=重要采样,
description={Importance Sampling},
sort={Importance Sampling},
}
\newglossaryentry{undirected_model}
{
name=无向模型,
description={undirected Model},
sort={undirected Model},
}
\newglossaryentry{partition_function}
{
name=配分函数,
description={Partition Function},
sort={Partition Function},
}
\newglossaryentry{law_of_large_numbers}
{
name=大数定理,
description={Law of large number},
sort={Law of large number},
}
\newglossaryentry{central_limit_theorem}
{
name=中心极限定理,
description={central limit theorem},
sort={central limit theorem},
}
\newglossaryentry{energy_based_model}
{
name=基于能量的模型,
description={Energy-based model},
symbol={EBM},
sort={Energy-based model},
}
\newglossaryentry{tempering}
{
name=回火,
description={tempering},
sort={tempering},
}
\newglossaryentry{biased_importance_sampling}
{
name=有偏重要采样,
description={biased importance sampling},
sort={biased importance sampling},
}
\newglossaryentry{VAE}
{
name=变分自编码器,
description={variational auto-encoder},
sort={variational auto-encoder},
symbol={VAE},
}
\newglossaryentry{CV}
{
name=计算机视觉,
description={Computer Vision},
sort={Computer Vision},
}
\newglossaryentry{SR}
{
name=语音识别,
description={Speech Recognition},
sort={Speech Recognition},
}
\newglossaryentry{NLP}
{
name=自然语言处理,
description={Natural Language Processing},
sort={Natural Language Processing},
symbol={NLP}
}
\newglossaryentry{RBM}
{
name=受限玻尔兹曼机,
description={Restricted Boltzmann Machine},
sort={Restricted Boltzmann Machine},
symbol={RBM}
}
\newglossaryentry{discriminative_RBM}
{
name=判别RBM,
description={discriminative RBM},
sort={discriminative RBM},
}
\newglossaryentry{Boltzmann}
{
name=玻尔兹曼,
description={Boltzmann},
sort={Boltzmann},
}
\newglossaryentry{BM}
{
name=玻尔兹曼机,
description={Boltzmann Machine},
sort={Boltzmann Machine},
}
\newglossaryentry{DBM}
{
name=深度玻尔兹曼机,
description={Deep Boltzmann Machine},
sort={Deep Boltzmann Machine},
symbol={DBM}
}
\newglossaryentry{CBM}
{
name=卷积玻尔兹曼机,
description={Convolutional Boltzmann Machine},
sort={Convolutional Boltzmann Machine},
symbol={CBM}
}
\newglossaryentry{directed_model}
{
name=有向模型,
description={Directed Model},
sort={Directed Model},
}
\newglossaryentry{ancestral_sampling}
{
name=原始采样,
description={Ancestral Sampling},
sort={Ancestral Sampling},
}
\newglossaryentry{stochastic_matrix}
{
name=随机矩阵,
description={Stochastic Matrix},
sort={Stochastic Matrix},
}
\newglossaryentry{stationary_distribution}
{
name=平稳分布,
description={Stationary Distribution},
sort={Stationary Distribution},
}
\newglossaryentry{equilibrium_distribution}
{
name=均衡分布,
description={Equilibrium Distribution},
sort={Equilibrium Distribution},
}
\newglossaryentry{index}
{
name=索引,
description={index of matrix},
sort={index of matrix},
}
\newglossaryentry{burn_in}
{
name=磨合,
description={Burning-in},
sort={Burning-in},
}
\newglossaryentry{mixing_time}
{
name=混合时间,
description={Mixing Time},
sort={Mixing Time},
}
\newglossaryentry{mixing}
{
name=混合,
description={Mixing},
sort={Mixing},
}
\newglossaryentry{gibbs_sampling}
{
name=Gibbs采样,
description={Gibbs Sampling},
sort={Gibbs Sampling},
}
\newglossaryentry{block_gibbs_sampling}
{
name=块吉布斯采样,
description={block Gibbs Sampling},
sort={block Gibbs Sampling},
}
\newglossaryentry{gibbs_steps}
{
name=吉布斯步数,
description={Gibbs steps},
sort={Gibbs steps},
}
\newglossaryentry{bagging}
{
name=Bagging,
description={bootstrap aggregating},
sort={bagging},
}
\newglossaryentry{mask}
{
name=掩码,
description={mask},
sort={mask},
}
\newglossaryentry{batch_normalization}
{
name=批标准化,
description={batch normalization},
sort={batch normalization},
}
\newglossaryentry{Batch_normalization}
{
name=批标准化,
description={Batch normalization},
sort={Batch normalization},
}
\newglossaryentry{parameter_sharing}
{
name=参数共享,
description={parameter sharing},
sort={parameter sharing},
}
\newglossaryentry{KL}
{
name=KL散度,
description={KL divergence},
sort={KL},
}
\newglossaryentry{temperature}
{
name=温度,
description={temperature},
sort={temperature},
}
\newglossaryentry{critical_temperatures}
{
name=临界温度,
description={critical temperatures},
sort={critical temperatures},
}
\newglossaryentry{parallel_tempering}
{
name=并行回火,
description={parallel tempering},
sort={parallel tempering},
}
\newglossaryentry{ASR}
{
name=自动语音识别,
description={Automatic Speech Recognition},
sort={Automatic Speech Recognition},
symbol={ASR}
}
\newglossaryentry{GP_GPU}
{
name=通用GPU,
description={general purpose GPU},
sort={general purpose GPU},
}
\newglossaryentry{coalesced}
{
name=级联,
description={coalesced},
sort={coalesced},
}
\newglossaryentry{warp}
{
name=warp,
description={warp},
sort={warp},
}
\newglossaryentry{data_parallelism}
{
name=数据并行,
description={data parallelism},
sort={data parallelism},
}
\newglossaryentry{model_parallelism}
{
name=模型并行,
description={model parallelism},
sort={model parallelism},
}
\newglossaryentry{ASGD}
{
name=异步随机梯度下降,
description={Asynchoronous Stochastic Gradient Descent},
sort={Asynchoronous Stochastic Gradient Descent},
}
\newglossaryentry{parameter_server}
{
name=参数服务器,
description={parameter server},
sort={parameter server},
}
\newglossaryentry{model_compression}
{
name=模型压缩,
description={model compression},
sort={model compression},
}
\newglossaryentry{dynamic_structure}
{
name=动态结构,
description={dynamic structure},
sort={dynamic structure},
}
\newglossaryentry{conditional_computation}
{
name=条件计算,
description={conditional computation},
sort={conditional computation},
}
\newglossaryentry{sphering}
{
name=sphering,
description={sphering},
sort={sphering},
}
\newglossaryentry{GCN}
{
name=全局对比度归一化,
description={Global contrast normalization},
sort={Global contrast normalization},
symbol={GCN}
}
\newglossaryentry{LCN}
{
name=局部对比度归一化,
description={local contrast normalization},
symbol={LCN},
sort={local contrast normalization},
}
\newglossaryentry{HMM}
{
name=隐马尔可夫模型,
description={Hidden Markov Model},
sort={Hidden Markov Model},
symbol={HMM}
}
\newglossaryentry{GMM}
{
name=高斯混合模型,
description={Gaussian Mixture Model},
sort={Gaussian Mixture Model},
symbol={GMM}
}
\newglossaryentry{transcribe}
{
name=转录,
description={transcribe},
sort={transcribe},
}
\newglossaryentry{PCA}
{
name=主成分分析,
description={principal components analysis},
sort={principal components analysis},
symbol={PCA}
}
\newglossaryentry{FA}
{
name=因子分析,
description={factor analysis},
sort={factor analysis},
}
\newglossaryentry{ICA}
{
name=独立成分分析,
description={independent component analysis},
sort={independent component analysis},
symbol={ICA}
}
\newglossaryentry{tICA}
{
name=地质ICA,
description={topographic ICA},
sort={topo independent component analysis},
}
\newglossaryentry{sparse_coding}
{
name=稀疏编码,
description={sparse coding},
sort={sparse coding},
}
\newglossaryentry{fixed_point_arithmetic}
{
name=定点运算,
description={fixed-point arithmetic},
sort={fixed-point arithmetic},
}
\newglossaryentry{float_point_arithmetic}
{
name=浮点运算,
description={float-point arithmetic},
sort={float-point arithmetic},
}
\newglossaryentry{GPU}
{
name=图形处理器,
description={Graphics Processing Unit},
sort={Graphics Processing Unit},
symbol={GPU}
}
\newglossaryentry{generative_model}
{
name=生成模型,
description={generative model},
sort={generative model},
}
\newglossaryentry{generative_modeling}
{
name=生成式建模,
description={generative modeling},
sort={generative modeling},
}
\newglossaryentry{dataset_augmentation}
{
name=数据集增强,
description={dataset augmentation},
sort={dataset augmentation},
}
\newglossaryentry{whitening}
{
name=白化,
description={whitening},
sort={whitening},
}
\newglossaryentry{DNN}
{
name=深度神经网络,
description={DNN},
sort={DNN},
}
\newglossaryentry{end_to_end}
{
name=端到端的,
description={end-to-end},
sort={end-to-end},
}
\newglossaryentry{structured_probabilistic_models}
{
name=结构化概率模型,
description={structured probabilistic model},
sort={structured probabilistic model},
}
\newglossaryentry{graphical_models}
{
name=图模型,
description={graphical model},
sort={graphical model},
}
\newglossaryentry{directed_graphical_model}
{
name=有向图模型,
description={directed graphical model},
sort={directed graphical model},
}
\newglossaryentry{dependency}
{
name=依赖,
description={dependency},
sort={dependency},
}
\newglossaryentry{bayesian_network}
{
name=贝叶斯网络,
description={Bayesian network},
sort={Bayesian network},
}
\newglossaryentry{model_averaging}
{
name=模型平均,
description={model averaging},
sort={model averaging},
}
\newglossaryentry{boosting}
{
name=Boosting,
description={Boosting},
sort={Boosting},
}
\newglossaryentry{weight_scaling_inference_rule}
{
name=权重比例推断规则,
description={weight scaling inference rule},
sort={weight scaling inference rule},
}
\newglossaryentry{statement}
{
name=声明,
description={statement},
sort={statement},
}
\newglossaryentry{quantum_mechanics}
{
name=量子力学,
description={quantum mechanics},
sort={quantum mechanics},
}
\newglossaryentry{subatomic}
{
name=亚原子,
description={subatomic},
sort={subatomic},
}
\newglossaryentry{fidelity}
{
name=逼真度,
description={fidelity},
sort={fidelity},
}
\newglossaryentry{degree_of_belief}
{
name=信任度,
description={degree of belief},
sort={degree of belief},
}
\newglossaryentry{frequentist_probability}
{
name=频率派概率,
description={frequentist probability},
sort={frequentist probability},
}
\newglossaryentry{subsample}
{
name=子采样,
description={subsample},
sort={subsample},
}
\newglossaryentry{bayesian_probability}
{
name=贝叶斯概率,
description={Bayesian probability},
sort={Bayesian probability},
}
\newglossaryentry{likelihood}
{
name=似然,
description={likelihood},
sort={likelihood},
}
\newglossaryentry{RV}
{
name=随机变量,
description={random variable},
sort={random variable},
}
\newglossaryentry{PD}
{
name=概率分布,
description={probability distribution},
sort={probability distribution},
}
\newglossaryentry{PMF}
{
name=概率质量函数,
description={probability mass function},
sort={probability mass function},
symbol={PMF}
}
\newglossaryentry{joint_probability_distribution}
{
name=联合概率分布,
description={joint probability distribution},
sort={joint probability distribution},
}
\newglossaryentry{normalized}
{
name=归一化的,
description={normalized},
sort={normalized},
}
\newglossaryentry{uniform_distribution}
{
name=均匀分布,
description={uniform distribution},
sort={uniform distribution},
}
\newglossaryentry{PDF}
{
name=概率密度函数,
description={probability density function},
sort={probability density function},
symbol={PDF}
}
\newglossaryentry{cumulative_function}
{
name=累积函数,
description={cumulative function},
sort={cumulative function},
}
\newglossaryentry{marginal_probability_distribution}
{
name=边缘概率分布,
description={marginal probability distribution},
sort={marginal probability distribution},
}
\newglossaryentry{sum_rule}
{
name=求和法则,
description={sum rule},
sort={sum rule},
}
\newglossaryentry{conditional_probability}
{
name=条件概率,
description={conditional probability},
sort={conditional probability},
}
\newglossaryentry{intervention_query}
{
name=干预查询,
description={intervention query},
sort={intervention query},
}
\newglossaryentry{causal_modeling}
{
name=因果模型,
description={causal modeling},
sort={causal modeling},
}
\newglossaryentry{causal_factor}
{
name=因果因子,
description={causal factor},
sort={causal factor},
}
\newglossaryentry{chain_rule}
{
name=链式法则,
description={chain rule},
sort={chain rule},
}
\newglossaryentry{product_rule}
{
name=乘法法则,
description={product rule},
sort={product rule},
}
\newglossaryentry{independent}
{
name=相互独立的,
description={independent},
sort={independent},
}
\newglossaryentry{conditionally_independent}
{
name=条件独立的,
description={conditionally independent},
sort={conditionally independent},
}
\newglossaryentry{expectation}
{
name=期望,
description={expectation},
sort={expectation},
}
\newglossaryentry{expected_value}
{
name=期望值,
description={expected value},
sort={expected value},
}
\newglossaryentry{example:chap5}
{
name=样本,
description={example},
sort={example},
}
\newglossaryentry{feature}
{
name=特征,
description={feature},
sort={feature},
}
\newglossaryentry{accuracy}
{
name=准确率,
description={accuracy},
sort={accuracy},
}
\newglossaryentry{error_rate}
{
name=错误率,
description={error rate},
sort={error rate},
}
\newglossaryentry{training_set}
{
name=训练集,
description={training set},
sort={training set},
}
\newglossaryentry{explanatory_factor}
{
name=解释因子,
description={explanatory factort},
sort={explanatory factor},
}
\newglossaryentry{underlying}
{
name=潜在,
description={underlying},
sort={underlying},
}
\newglossaryentry{underlying_cause}
{
name=潜在成因,
description={underlying cause},
sort={underlying cause},
}
\newglossaryentry{test_set}
{
name=测试集,
description={test set},
sort={test set},
}
\newglossaryentry{performance_measures}
{
name=性能度量,
description={performance measures},
sort={performance measures},
}
\newglossaryentry{experience}
{
name=经验,
description={experience, E},
sort={experience, E},
}
\newglossaryentry{unsupervised}
{
name=无监督,
description={unsupervised},
sort={unsupervised},
}
\newglossaryentry{supervised}
{
name=监督,
description={supervised},
sort={supervised},
}
\newglossaryentry{semi_supervised}
{
name=半监督,
description={semi-supervised},
sort={semi-supervised},
}
\newglossaryentry{supervised_learning}
{
name=监督学习,
description={supervised learning},
sort={supervised learning},
}
\newglossaryentry{unsupervised_learning}
{
name=无监督学习,
description={unsupervised learning},
sort={unsupervised learning},
}
\newglossaryentry{dataset}
{
name=数据集,
description={dataset},
sort={dataset},
}
\newglossaryentry{setting}
{
name=情景,
description={setting},
sort={setting},
}
\newglossaryentry{data_points}
{
name=数据点,
description={data point},
sort={data point},
}
\newglossaryentry{label}
{
name=标签,
description={label},
sort={label},
}
\newglossaryentry{labeled}
{
name=标注,
description={labeled},
sort={labeled},
}
\newglossaryentry{unlabeled}
{
name=未标注,
description={unlabeled},
sort={unlabeled},
}
\newglossaryentry{target}
{
name=目标,
description={target},
sort={target},
}
\newglossaryentry{reinforcement_learning}
{
name=强化学习,
description={reinforcement learning},
sort={reinforcement learning},
}
\newglossaryentry{design_matrix}
{
name=设计矩阵,
description={design matrix},
sort={design matrix},
}
\newglossaryentry{parameters}
{
name=参数,
description={parameter},
sort={parameter},
}
\newglossaryentry{weights}
{
name=权重,
description={weight},
sort={weight},
}
\newglossaryentry{mean_squared_error}
{
name=均方误差,
description={mean squared error},
sort={mean squared error},
symbol={MSE}
}
\newglossaryentry{normal_equations}
{
name=正规方程,
description={normal equation},
sort={normal equation},
}
\newglossaryentry{training_error}
{
name=训练误差,
description={training error},
sort={training error},
}
\newglossaryentry{generalization_error}
{
name=泛化误差,
description={generalization error},
sort={generalization error},
}
\newglossaryentry{test_error}
{
name=测试误差,
description={test error},
sort={test error},
}
\newglossaryentry{hypothesis_space}
{
name=假设空间,
description={hypothesis space},
sort={hypothesis space},
}
\newglossaryentry{capacity}
{
name=容量,
description={capacity},
sort={capacity},
}
\newglossaryentry{representational_capacity}
{
name=表示容量,
description={representational capacity},
sort={representational capacity},
}
\newglossaryentry{effective_capacity}
{
name=有效容量,
description={effective capacity},
sort={effective capacity},
}
\newglossaryentry{linear_threshold_units}
{
name=线性阈值单元,
description={linear threshold units},
sort={linear threshold units},
}
\newglossaryentry{nonparametric}
{
name=非参数,
description={non-parametric},
sort={non-parametric},
}
\newglossaryentry{nearest_neighbor_regression}
{
name=最近邻回归,
description={nearest neighbor regression},
sort={nearest neighbor regression},
}
\newglossaryentry{nearest_neighbor}
{
name=最近邻,
description={nearest neighbor},
sort={nearest neighbor},
}
\newglossaryentry{bayes_error}
{
name=贝叶斯误差,
description={Bayes error},
sort={Bayes error},
}
\newglossaryentry{no_free_lunch_theorem}
{
name=没有免费午餐定理,
description={no free lunch theorem},
sort={no free lunch theorem},
}
\newglossaryentry{validation_set}
{
name=验证集,
description={validation set},
sort={validation set},
}
\newglossaryentry{benchmarks}
{
name=基准,
description={bechmark},
sort={bechmark},
}
\newglossaryentry{baseline}
{
name=基准,
description={baseline},
sort={beseline}
}
\newglossaryentry{point_estimator}
{
name=点估计,
description={point estimator},
sort={point estimator},
}
\newglossaryentry{estimator:chap5}
{
name=估计量,
description={estimator},
sort={estimator},
}
\newglossaryentry{statistics}
{
name=统计量,
description={statistics},
sort={statistics},
}
\newglossaryentry{unbiased}
{
name=无偏,
description={unbiased},
sort={unbiased},
}
\newglossaryentry{biased}
{
name=有偏,
description={biased},
sort={biased},
}
\newglossaryentry{asynchronous}
{
name=异步,
description={asynchronous},
sort={asynchronous},
}
\newglossaryentry{asymptotically_unbiased}
{
name=渐近无偏,
description={asymptotically unbiased},
sort={asymptotically unbiased},
}
\newglossaryentry{sample_mean}
{
name=样本均值, %采样均值?
description={sample mean},
sort={sample mean},
}
\newglossaryentry{sample_variance}
{
name=样本方差, %采样方差?
description={sample variance},
sort={sample variance},
}
\newglossaryentry{unbiased_sample_variance}
{
name=无偏样本方差, %采样方差?
description={unbiased sample variance},
sort={unbiased sample variance},
}
\newglossaryentry{standard_error}
{
name=标准差,
description={standard error},
symbol={SE},
sort={standard error},
}
\newglossaryentry{consistency}
{
name=一致性,
description={consistency},
sort={consistency},
}
\newglossaryentry{almost_sure}
{
name=几乎必然,
description={almost sure},
sort={almost sure},
}
\newglossaryentry{almost_sure_convergence}
{
name=几乎必然收敛,
description={almost sure convergence},
sort={almost sure convergence},
}
\newglossaryentry{statistical_efficiency}
{
name=统计效率,
description={statistic efficiency},
sort={statistic efficiency},
}
\newglossaryentry{parametric_case}
{
name=有参情况,
description={parametric case},
sort={parametric case},
}
\newglossaryentry{frequentist_statistics}
{
name=频率派统计,
description={frequentist statistics},
sort={frequentist statistics},
}
\newglossaryentry{bayesian_statistics}
{
name=贝叶斯统计,
description={Bayesian statistics},
sort={Bayesian statistics},
}
\newglossaryentry{prior_probability_distribution}
{
name=先验概率分布,
description={prior probability distribution},
sort={prior probability distribution},
}
\newglossaryentry{maximum_a_posteriori}
{
name=最大后验,
description={maximum a posteriori},
sort={maximum a posteriori},
}
\newglossaryentry{maximum_likelihood_estimation}
{
name=最大似然估计,
description={maximum likelihood estimation},
sort={maximum likelihood estimation},
}
\newglossaryentry{maximum_likelihood}
{
name=最大似然,
description={maximum likelihood},
sort={maximum likelihood},
}
\newglossaryentry{kernel_trick}
{
name=核技巧,
description={kernel trick},
sort={kernel trick},
}
\newglossaryentry{kernel}
{
name=核函数,
description={kernel function},
sort={kernel function},
}
\newglossaryentry{gaussian_kernel}
{
name=高斯核,
description={Gaussian kernel},
sort={Gaussian kernel},
}
\newglossaryentry{kernel_machines}
{
name=核机器,
description={kernel machine},
sort={kernel machine},
}
\newglossaryentry{kernel_methods}
{
name=核方法,
description={kernel method},
sort={kernel method},
}
\newglossaryentry{support_vectors}
{
name=支持向量,
description={support vector},
sort={support vector},
}
\newglossaryentry{SVM}
{
name=支持向量机,
description={support vector machine},
symbol={SVM}
}
\newglossaryentry{phoneme}
{
name=音素,
description={phoneme},
sort={phoneme},
}
\newglossaryentry{acoustic}
{
name=声学,
description={acoustic},
sort={acoustic},
}
\newglossaryentry{phonetic}
{
name=语音,
description={phonetic},
sort={phonetic},
}
\newglossaryentry{mixture_of_experts}
{
name=专家混合体,
description={mixture of experts},
sort={mixture of experts},
}
\newglossaryentry{gauss_mixture}
{
name=高斯混合体,
description={Gaussian mixtures},
sort={Gaussian mixtures},
}
\newglossaryentry{hard_mixture_of_experts}
{
name=硬专家混合体,
description={hard mixture of experts},
sort={hard mixture of experts},
}
\newglossaryentry{gater}
{
name=选通器,
description={gater},
sort={gater},
}
\newglossaryentry{expert_network}
{
name=专家网络,
description={expert network},
sort={expert network},
}
\newglossaryentry{attention_mechanism}
{
name=注意力机制,
description={attention mechanism},
sort={attention mechanism},
}
\newglossaryentry{fast_dropout}
{
name=快速Dropout,
description={fast dropout},
sort={fast dropout},
}
\newglossaryentry{dropout_boosting}
{
name=Dropout Boosting,
description={Dropout Boosting},
sort={dropout boosting},
}
\newglossaryentry{adversarial_example}
{
name=对抗样本,
description={adversarial example},
sort={adversarial example},
}
\newglossaryentry{adversarial}
{
name=对抗,
description={adversarial},
sort={adversarial},
}
\newglossaryentry{virtual_adversarial_example}
{
name=虚拟对抗样本,
description={virtual adversarial example},
sort={virtual adversarial example},
}
\newglossaryentry{adversarial_training}
{
name=对抗训练,
description={adversarial training},
sort={adversarial training},
}
\newglossaryentry{virtual_adversarial_training}
{
name=虚拟对抗训练,
description={virtual adversarial training},
sort={virtual adversarial training}
}
\newglossaryentry{tangent_distance}
{
name=切面距离,
description={tangent distance},
sort={tangent distance},
}
\newglossaryentry{tangent_prop}
{
name=正切传播,
description={tangent prop},
sort={tangent prop},
}
\newglossaryentry{tangent_propagation}
{
name=正切传播,
description={tangent propagation}
}
\newglossaryentry{double_backprop}
{
name=双反向传播,
description={double backprop},
sort={double backprop},
}
\newglossaryentry{EM}
{
name=期望最大化,
description={expectation maximization},
sort={expectation maximization},
symbol={EM}
}
\newglossaryentry{mean_field}
{
name=均值场,
description={mean-field},
sort={mean-field},
}
\newglossaryentry{ELBO}
{
name=证据下界,
description={evidence lower bound},
sort={evidence lower bound},
symbol={ELBO}
}
\newglossaryentry{variational_free_energy}
{
name=变分自由能,
description={variational free energy},
sort={variational free energy},
}
\newglossaryentry{structured_variational_inference}
{
name=结构化变分推断,
description={structured variational inference},
sort={structured variational inference},
}
\newglossaryentry{variational_inference}
{
name=变分推断,
description={variational inference},
sort={variational inference},
}
\newglossaryentry{binary_sparse_coding}
{
name=二值稀疏编码,
description={binary sparse coding},
sort={binary sparse coding},
}
\newglossaryentry{feedforward_network}
{
name=前馈网络,
description={feedforward network},
sort={feedforward network},
}
\newglossaryentry{transition}
{
name=转移,
description={transition},
sort={transition},
}
\newglossaryentry{reconstruction}
{
name=重构,
description={reconstruction},
sort={reconstruction},
}
\newglossaryentry{GSN}
{
name=生成随机网络,
description={generative stochastic network},
sort={generative stochastic network},
symbol={GSN}
}
\newglossaryentry{score_matching}
{
name=得分匹配,
description={score matching},
sort={score matching},
}
\newglossaryentry{factorial}
{
name=因子,
description={factorial},
sort={factorial},
}
\newglossaryentry{factorized}
{
name=分解的,
description={factorized},
sort={factorized},
}
\newglossaryentry{meanfield}
{
name=均匀场,
description={meanfield},
sort={meanfield},
}
\newglossaryentry{MLE}
{
name=最大似然估计,
description={maximum likelihood estimation},
sort={maximum likelihood estimation},
}
\newglossaryentry{PPCA}
{
name=概率PCA,
description={probabilistic PCA},
sort={probabilistic PCA},
symbol={PPCA}
}
\newglossaryentry{SGA}
{
name=随机梯度上升,
description={Stochastic Gradient Ascent},
sort={Stochastic Gradient Ascent},
}
\newglossaryentry{clique}
{
name=团,
description={clique},
sort={clique},
}
\newglossaryentry{dirac_distribution}
{
name=Dirac分布,
description={dirac distribution},
sort={dirac distribution},
}
\newglossaryentry{fixed_point_equation}
{
name=不动点方程,
description={fixed point equation},
sort={fixed point equation},
}
\newglossaryentry{calculus_of_variations}
{
name=变分法,
description={calculus of variations},
sort={calculus of variations},
}
\newglossaryentry{wake_sleep}
{
name=醒眠,
description={wake sleep},
sort={wake sleep},
}
\newglossaryentry{BN}
{
name=信念网络,
description={belief network},
sort={belief network},
}
\newglossaryentry{MRF}
{
name=马尔可夫随机场,
description={Markov random field},
sort={Markov random field},
symbol={MRF}
}
\newglossaryentry{markov_network}
{
name=马尔可夫网络,
description={Markov network},
sort={Markov network},
}
\newglossaryentry{log_linear_model}
{
name=对数线性模型,
description={log-linear model},
sort={log-linear model},
}
\newglossaryentry{product_of_expert}
{
name=专家之积,
description={product of expert},
sort={product of expert},
}
\newglossaryentry{free_energy}
{
name=自由能,
description={free energy},
sort={free energy},
}
\newglossaryentry{harmony}
{
name=harmony,
description={harmony},
sort={harmony},
}
\newglossaryentry{separation}
{
name=分离,
description={separation},
sort={separation},
}
\newglossaryentry{separate}
{
name=分离的,
description={separate},
sort={separate},
}
\newglossaryentry{dseparation}
{
name=d-分离,
description={d-separation},
sort={d-separation},
}
\newglossaryentry{local_conditional_probability_distribution}
{
name=局部条件概率分布,
description={local conditional probability distribution},
sort={local conditional probability distribution},
}
\newglossaryentry{conditional_probability_distribution}
{
name=条件概率分布,
description={conditional probability distribution}
}
\newglossaryentry{boltzmann_distribution}
{
name=玻尔兹曼分布,
description={Boltzmann distribution},
sort={Boltzmann distribution},
}
\newglossaryentry{gibbs_distribution}
{
name=吉布斯分布,
description={Gibbs distribution},
sort={Gibbs distribution},
}
\newglossaryentry{energy_function}
{
name=能量函数,
description={energy function},
sort={energy function},
}
\newglossaryentry{immorality}
{
name=不道德,
description={immorality},
sort={immorality},
}
\newglossaryentry{moralization}
{
name=道德化,
description={moralization},
sort={moralization},
}
\newglossaryentry{moralized_graph}
{
name=道德图,
description={moralized graph},
sort={moralized graph},
}
\newglossaryentry{standard_deviation}
{
name=标准差,
description={standard deviation},
symbol={SD},
sort={standard deviation},
}
\newglossaryentry{correlation}
{
name=相关系数,
description={correlation},
sort={correlation},
}
\newglossaryentry{standard_normal_distribution}
{
name=标准正态分布,
description={standard normal distribution},
sort={standard normal distribution},
}
\newglossaryentry{covariance_matrix}
{
name=协方差矩阵,
description={covariance matrix},
sort={covariance matrix},
}
\newglossaryentry{bernoulli_distribution}
{
name=Bernoulli分布,
description={Bernoulli distribution},
sort={Bernoulli distribution},
}
\newglossaryentry{bernoulli_output_distribution}
{
name=Bernoulli输出分布,
description={Bernoulli output distribution},
sort={Bernoulli output distribution},
}
\newglossaryentry{multinoulli_distribution}
{
name=Multinoulli分布,
description={multinoulli distribution},
sort={multinoulli distribution},
}
\newglossaryentry{multinoulli_output_distribution}
{
name=Multinoulli输出分布,
description={multinoulli output distribution},
sort={multinoulli output distribution},
}
\newglossaryentry{categorical_distribution}
{
name=范畴分布,
description={categorical distribution},
sort={categorical distribution},
}
\newglossaryentry{multinomial_distribution}
{
name=多项式分布,
description={multinomial distribution},
sort={multinomial distribution},
}
\newglossaryentry{normal_distribution}
{
name=正态分布,
description={normal distribution},
sort={normal distribution},
}
\newglossaryentry{gaussian_distribution}
{
name=高斯分布,
description={Gaussian distribution},
sort={Gaussian distribution},
}
\newglossaryentry{precision}
{
name=精度,
description={precision},
sort={precision},
}
\newglossaryentry{multivariate_normal_distribution}
{
name=多维正态分布,
description={multivariate normal distribution},
sort={multivariate normal distribution},
}
\newglossaryentry{precision_matrix}
{
name=精度矩阵,
description={precision matrix},
sort={precision matrix},
}
\newglossaryentry{isotropic}
{
name=各向同性,
description={isotropic},
sort={isotropic},
}
\newglossaryentry{exponential_distribution}
{
name=指数分布,
description={exponential distribution},
sort={exponential distribution},
}
\newglossaryentry{indicator_function}
{
name=指示函数,
description={indicator function},
sort={indicator function},
}
\newglossaryentry{laplace_distribution}
{
name=Laplace分布,
description={Laplace distribution},
sort={Laplace distribution},
}
\newglossaryentry{dirac_delta_function}
{
name=Dirac delta函数,
description={Dirac delta function},
sort={Dirac delta function},
}
\newglossaryentry{generalized_function}
{
name=广义函数,
description={generalized function},
sort={generalized function},
}
\newglossaryentry{empirical_distribution}
{
name=经验分布,
description={empirical distribution},
sort={empirical distribution},
}
\newglossaryentry{empirical_frequency}
{
name=经验频率,
description={empirical frequency},
sort={empirical frequency},
}
\newglossaryentry{mixture_distribution}
{
name=混合分布,
description={mixture distribution},
sort={mixture distribution},
}
\newglossaryentry{latent_variable}
{
name=潜变量,
description={latent variable},
sort={latent variable},
}
\newglossaryentry{hidden_variable}
{
name=隐藏变量,
description={hidden variable},
sort={hidden variable},
}
\newglossaryentry{prior_probability}
{
name=先验概率,
description={prior probability},
sort={prior probability},
}
\newglossaryentry{posterior_probability}
{
name=后验概率,
description={posterior probability},
sort={posterior probability},
}
\newglossaryentry{universal_approximator}
{
name=万能近似器,
description={universal approximator},
sort={universal approximator},
}
\newglossaryentry{universal_function_approximator}
{
name=万能函数近似器,
description={universal function approximator},
sort={universal function approximator},
}
\newglossaryentry{logistic_sigmoid}
{
name=logistic sigmoid,
description={logistic sigmoid},
sort={logistic sigmoid},
}
\newglossaryentry{sigmoid}
{
name=sigmoid,
description={sigmoid},
sort={sigmoid},
}
\newglossaryentry{saturate}
{
name=饱和,
description={saturate},
sort={saturate},
}
\newglossaryentry{softplus_function}
{
name=softplus函数,
description={softplus function},
sort={softplus function},
}
\newglossaryentry{logit}
{
name=分对数,
description={logit},
sort={logit},
}
\newglossaryentry{positive_part_function}
{
name=正部函数,
description={positive part function},
sort={positive part function},
}
\newglossaryentry{negative part function}
{
name=负部函数,
description={negative part function},
sort={negative part function},
}
\newglossaryentry{bayes_rule}
{
name=贝叶斯规则,
description={Bayes' rule},
sort={Bayes' rule},
}
\newglossaryentry{measure_theory}
{
name=测度论,
description={measure theory},
sort={measure theory},
}
\newglossaryentry{measure_zero}
{
name=零测度,
description={measure zero},
sort={measure zero},
}
\newglossaryentry{almost_everywhere}
{
name=几乎处处,
description={almost everywhere},
sort={almost everywhere},
}
\newglossaryentry{jacobian_matrix}
{
name=Jacobian矩阵,
description={Jacobian matrix},
sort={Jacobian matrix},
}
\newglossaryentry{self_information}
{
name=自信息,
description={self-information},
sort={self-information},
}
\newglossaryentry{nats}
{
name=奈特,
description={nats},
sort={nats},
}
\newglossaryentry{bits}
{
name=比特,
description={bit},
sort={bit},
}
\newglossaryentry{shannons}
{
name=香农,
description={shannons},
sort={shannons},
}
\newglossaryentry{Shannon_entropy}
{
name=香农熵,
description={Shannon entropy},
sort={Shannon entropy},
}
\newglossaryentry{differential_entropy}
{
name=微分熵,
description={differential entropy},
sort={differential entropy},
}
\newglossaryentry{differential_equation}
{
name=微分方程,
description={differential equation},
sort={differential equation},
}
\newglossaryentry{KL_divergence}
{
name=KL散度,
description={Kullback-Leibler (KL) divergence},
sort={Kullback-Leibler (KL) divergence},
}
\newglossaryentry{cross_entropy}
{
name=交叉熵,
description={cross-entropy},
sort={cross-entropy},
}
\newglossaryentry{entropy}
{
name=熵,
description={entropy},
sort={entropy},
}
\newglossaryentry{factorization}
{
name=分解,
description={factorization},
sort={factorization},
}
\newglossaryentry{structured_probabilistic_model}
{
name=结构化概率模型,
description={structured probabilistic model},
sort={structured probabilistic model},
}
\newglossaryentry{graphical_model}
{
name=图模型,
description={graphical model},
sort={graphical model},
}
\newglossaryentry{backoff}
{
name=回退,
description={back-off},
sort={back-off},
}
\newglossaryentry{directed}
{
name=有向,
description={directed},
sort={directed},
}
\newglossaryentry{undirected}
{
name=无向,
description={undirected},
sort={undirected},
}
\newglossaryentry{undirected_graphical_model}
{
name=无向图模型,
description={undirected graphical model}
}
\newglossaryentry{proportional}
{
name=成比例,
description={proportional},
sort={proportional},
}
\newglossaryentry{description}
{
name=描述,
description={description},
sort={description},
}
\newglossaryentry{decision_tree}
{
name=决策树,
description={decision tree},
sort={decision tree},
}
\newglossaryentry{factor_graph}
{
name=因子图,
description={factor graph},
sort={factor graph},
}
\newglossaryentry{structure_learning}
{
name=结构学习,
description={structure learning},
sort={structure learning},
}
\newglossaryentry{loopy_belief_propagation}
{
name=环状信念传播,
description={loopy belief propagation},
sort={loopy belief propagation},
}
\newglossaryentry{harmonium}
{
name=簧风琴,
description={harmonium},
sort={harmonium},
}
\newglossaryentry{convolutional_network}
{
name=卷积网络,
description={convolutional network},
sort={convolutional network},
}
\newglossaryentry{convolutional_net}
{
name=卷积网络,
description={convolutional net},
sort={convolutional net},
}
\newglossaryentry{main_diagonal}
{
name=主对角线,
description={main diagonal},
sort={main diagonal},
}
\newglossaryentry{transpose}
{
name=转置,
description={transpose},
sort={transpose},
}
\newglossaryentry{broadcasting}
{
name=广播,
description={broadcasting},
sort={broadcasting},
}
\newglossaryentry{matrix_product}
{
name=矩阵乘积,
description={matrix product},
sort={matrix product},
}
\newglossaryentry{adagrad}
{
name=AdaGrad,
description={AdaGrad},
sort={AdaGrad},
}
\newglossaryentry{element_wise_product}
{
name=元素对应乘积,
description={element-wise product},
sort={element-wise product},
}
\newglossaryentry{hadamard_product}
{
name=Hadamard乘积,
description={Hadamard product},
sort={Hadamard product},
}
\newglossaryentry{clique_potential}
{
name=团势能,
description={clique potential},
sort={clique potential},
}
\newglossaryentry{factor}
{
name=因子,
description={factor},
sort={factor},
}
\newglossaryentry{unnormalized_probability_function}
{
name=未归一化概率函数,
description={unnormalized probability function},
sort={unnormalized probability function},
}
\newglossaryentry{recurrent_network}
{
name=循环网络,
description={recurrent network},
sort={recurrent network},
}
\newglossaryentry{vanish_explode_gradient}
{
name=梯度消失与爆炸问题,
description={vanishing and exploding gradient problem},
sort={vanishing and exploding gradient problem },
}
\newglossaryentry{vanish_gradient}
{
name=梯度消失,
description={vanishing gradient},
sort={vanishing gradient},
}
\newglossaryentry{explode_gradient}
{
name=梯度爆炸,
description={exploding gradient},
sort={exploding gradient},
}
\newglossaryentry{computational_graph}
{
name=计算图,
description={computational graph},
sort={computational graph},
}
\newglossaryentry{unfolding}
{
name=展开,
description={unfolding},
sort={unfolding},
}
\newglossaryentry{invert}
{
name=求逆,
description={invert},
sort={invert},
}
\newglossaryentry{time_step}
{
name=时间步,
description={time step},
sort={time step},
}
\newglossaryentry{n_gram}
{
name=$n$-gram,
description={n-gram},
sort={n-gram},
}
\newglossaryentry{curse_of_dimensionality}
{
name=维数灾难,
description={curse of dimensionality},
sort={curse of dimensionality},
}
\newglossaryentry{smoothness_prior}
{
name=平滑先验,
description={smoothness prior},
sort={smoothness prior},
}
\newglossaryentry{local_constancy_prior}
{
name=局部不变性先验,
description={local constancy prior},
sort={local constancy prior},
}
\newglossaryentry{local_kernel}
{
name=局部核,
description={local kernel},
sort={local kernel},
}
\newglossaryentry{manifold}
{
name=流形,
description={manifold},
sort={manifold},
}
\newglossaryentry{manifold_tangent_classifier}
{
name=流形正切分类器,
description={manifold tangent classifier}
}
\newglossaryentry{manifold_learning}
{
name=流形学习,
description={manifold learning},
sort={manifold learning},
}
\newglossaryentry{manifold_hypothesis}
{
name=流形假设,
description={manifold hypothesis},
sort={manifold hypothesis},
}
\newglossaryentry{loop}
{
name=环,
description={loop},
sort={loop},
}
\newglossaryentry{chord}
{
name=弦,
description={chord},
sort={chord},
}
\newglossaryentry{chordal_graph}
{
name=弦图,
description={chordal graph},
sort={chordal graph},
}
\newglossaryentry{triangulated_graph}
{
name=三角形化图,
description={triangulated graph},
sort={triangulated graph},
}
\newglossaryentry{triangulate}
{
name=三角形化,
description={triangulate},
sort={triangulate},
}
\newglossaryentry{risk}
{
name=风险,
description={risk},
sort={risk},
}
\newglossaryentry{empirical_risk}
{
name=经验风险,
description={empirical risk},
sort={empirical risk},
}
\newglossaryentry{empirical_risk_minimization}
{
name=经验风险最小化,
description={empirical risk minimization},
sort={empirical risk minimization},
}
\newglossaryentry{surrogate_loss_function}
{
name=代理损失函数,
description={surrogate loss function},
sort={surrogate loss function},
}
\newglossaryentry{batch}
{
name=批量,
description={batch},
sort={batch},
}
\newglossaryentry{deterministic}
{
name=确定性,
description={deterministic},
sort={deterministic},
}
\newglossaryentry{stochastic}
{
name=随机,
description={stochastic},
sort={stochastic},
}
\newglossaryentry{online}
{
name=在线,
description={online},
sort={online},
}
\newglossaryentry{minibatch_stochastic}
{
name=小批量随机, %小批量
description={minibatch stochastic},
sort={minibatch stochastic},
}
\newglossaryentry{stream}
{
name=流,
description={stream},
sort={stream},
}
\newglossaryentry{model_identifiability}
{
name=模型可辨识性,
description={model identifiability},
sort={model identifiability},
}
\newglossaryentry{weight_space_symmetry}
{
name=权重空间对称性,
description={weight space symmetry},
sort={weight space symmetry},
}
\newglossaryentry{saddle_free_newton_method}
{
name=无鞍牛顿法,
description={saddle-free Newton method},
sort={saddle-free Newton method},
}
\newglossaryentry{gradient_clipping}
{
name=梯度截断,
description={gradient clipping},
sort={gradient clipping},
}
\newglossaryentry{power_method}
{
name=幂方法,
description={power method},
sort={power method},
}
\newglossaryentry{linear_factor}
{
name=线性因子模型,
description={linear factor model},
sort={linear factor model},
}
\newglossaryentry{forward_propagation}
{
name=前向传播,
description={forward propagation},
sort={forward propagation},
}
\newglossaryentry{backward_propagation}
{
name=反向传播,
description={backward propagation},
sort={backward propagation},
}
\newglossaryentry{unfolded_graph}
{
name=展开图,
description={unfolded graph},
sort={unfolded graph},
}
\newglossaryentry{BPTT}
{
name=通过时间反向传播,
description={back-propagation through time},
sort={back-propagation through time},
symbol={BPTT}
}
\newglossaryentry{teacher_forcing}
{
name=导师驱动过程,
description={teacher forcing},
sort={teacher forcing},
}
\newglossaryentry{stationary}
{
name=平稳的,
description={stationary},
sort={stationary},
}
\newglossaryentry{deep_feedforward_network}
{
name=深度前馈网络,
description={deep feedforward network},
sort={deep feedforward network},
}
\newglossaryentry{feedforward_neural_network}
{
name=前馈神经网络,
description={feedforward neural network},
sort={feedforward neural network},
}
\newglossaryentry{feedforward}
{
name=前向,
description={feedforward},
sort={feedforward},
}
\newglossaryentry{feedback}
{
name=反馈,
description={feedback},
sort={feedback},
}
\newglossaryentry{network}
{
name=网络,
description={network},
sort={network},
}
\newglossaryentry{first_layer}
{
name=第一层,
description={first layer},
sort={first layer},
}
\newglossaryentry{second_layer}
{
name=第二层,
description={second layer},
sort={second layer},
}
\newglossaryentry{depth}
{
name=深度,
description={depth},
sort={depth},
}
\newglossaryentry{output_layer}
{
name=输出层,
description={output layer},
sort={output layer},
}
\newglossaryentry{hidden_layer}
{
name=隐藏层,
description={hidden layer},
sort={hidden layer},
}
\newglossaryentry{width}
{
name=宽度,
description={width},
sort={width},
}
\newglossaryentry{unit}
{
name=单元,
description={unit},
sort={unit},
}
\newglossaryentry{activation_function}
{
name=激活函数,
description={activation function},
sort={activation function},
}
\newglossaryentry{dbd}
{
name=delta-bar-delta,
description={delta-bar-delta},
sort={delta-bar-delta},
}
\newglossaryentry{gaussian_output_distribution}
{
name=高斯输出分布,
description={Gaussian output distribution},
sort={Gaussian output distribution},
}
\newglossaryentry{back_propagation}
{
name=反向传播,
description={back propagation},
sort={back propagation},
}
\newglossaryentry{back_propagate}
{
name=反向传播,
description={back propagate},
sort={back propagate},
}
\newglossaryentry{BP}
{
name=反向传播,
description={backprop},
sort={backprop},
symbol={BP}
}
\newglossaryentry{functional}
{
name=泛函,
description={functional},
sort={functional},
}
\newglossaryentry{mean_absolute_error}
{
name=平均绝对误差,
description={mean absolute error},
sort={mean absolute error},
}
\newglossaryentry{winner_take_all}
{
name=赢者通吃,
description={winner-take-all},
sort={winner-take-all},
}
\newglossaryentry{heteroscedastic}
{
name=异方差,
description={heteroscedastic},
sort={heteroscedastic},
}
\newglossaryentry{mixture_density_network}
{
name=混合密度网络,
description={mixture density network},
sort={mixture density network},
}
\newglossaryentry{clip_gradients}
{
name=梯度截断,
description={clip gradient},
sort={clip gradient},
}
\newglossaryentry{absolute_value_rectification}
{
name=绝对值整流,
description={absolute value rectification},
sort={absolute value rectification},
}
\newglossaryentry{leaky_ReLU} %有没有更好的翻译
{
name=渗漏整流线性单元,
description={Leaky ReLU},
sort={Leaky ReLU},
symbol={Leaky ReLU}
}
\newglossaryentry{PReLU}
{
name=参数化整流线性单元,
description={parametric ReLU},
sort={parametric ReLU},
symbol={PReLU}
}
\newglossaryentry{maxout_unit}
{
name=maxout单元,
description={maxout unit},
sort={maxout unit},
}
\newglossaryentry{maxout}
{
name=maxout,
description={maxout},
symbol={maxout}
}
\newglossaryentry{catastrophic_forgetting}
{
name=灾难遗忘,
description={catastrophic forgetting},
sort={catastrophic forgetting},
}
\newglossaryentry{RBF}
{
name=径向基函数,
description={radial basis function},
sort={radial basis function},
symbol={RBF}
}
\newglossaryentry{softplus}
{
name=softplus,
description={softplus},
sort={softplus},
}
\newglossaryentry{hard_tanh}
{
name=硬双曲正切函数,
description={hard tanh},
sort={hard tanh},
}
\newglossaryentry{architecture}
{
name=架构,
description={architecture},
sort={architecture},
}
\newglossaryentry{universal_approximation_theorem}
{
name=万能近似定理, %万能逼近定理?
description={universal approximation theorem},
sort={universal approximation theorem},
}
\newglossaryentry{operation}
{
name=操作,
description={operation},
sort={operation},
}
\newglossaryentry{symbol}
{
name=符号,
description={symbol},
sort={symbol},
}
\newglossaryentry{numeric_value}
{
name=数值,
description={numeric value},
sort={numeric value},
}
\newglossaryentry{dynamic_programming}
{
name=动态规划,
description={dynamic programming},
sort={dynamic programming},
}
\newglossaryentry{automatic_differentiation}
{
name=自动微分,
description={automatic differentiation},
sort={automatic differentiation},
}
\newglossaryentry{reverse_mode_accumulation}
{
name=反向模式累加,
description={reverse mode accumulation},
sort={reverse mode accumulation},
}
\newglossaryentry{forward_mode_accumulation}
{
name=前向模式累加,
description={forward mode accumulation},
sort={forward mode accumulation},
}
\newglossaryentry{Krylov_methods}
{
name=Krylov方法,
description={Krylov method},
sort={Krylov method},
}
\newglossaryentry{parallel_distributed_processing}
{
name=并行分布式处理,
description={Parallel Distributed Processing},
sort={Parallel Distributed Processing},
}
\newglossaryentry{sparse_activation}
{
name=稀疏激活,
description={sparse activation},
sort={sparse activation},
}
\newglossaryentry{damping}
{
name=衰减,
description={damping},
sort={damping},
}
\newglossaryentry{learned}
{
name=学成,
description={learned},
sort={learned},
}
\newglossaryentry{message_passing}
{
name=信息传输,
description={message passing},
sort={message passing},
}
\newglossaryentry{functional_derivative}
{
name=泛函导数,
description={functional derivative},
sort={functional derivative},
}
\newglossaryentry{variational_derivative}
{
name=变分导数,
description={variational derivative},
sort={variational derivative},
}
\newglossaryentry{excess_error}
{
name=额外误差,
description={excess error},
sort={excess error},
}
\newglossaryentry{momentum}
{
name=动量,
description={momentum},
sort={momentum},
}
\newglossaryentry{nmomentum}
{
name=Nesterov 动量,
description={Nesterov momentum},
sort={Nesterov momentum},
}
\newglossaryentry{chaos}
{
name=混沌,
description={chaos},
sort={chaos},
}
\newglossaryentry{normalized_initialization}
{
name=标准初始化, % ??
description={normalized initialization},
sort={normalized initialization},
}
\newglossaryentry{sparse_initialization}
{
name=稀疏初始化,
description={sparse initialization},
sort={sparse initialization},
}
\newglossaryentry{conjugate_directions}
{
name=共轭方向,
description={conjugate directions},
sort={conjugate directions},
}
\newglossaryentry{conjugate}
{
name=共轭,
description={conjugate},
sort={conjugate},
}
\newglossaryentry{conditional_independent}
{
name=条件独立,
description={conditionally independent},
sort={conditionally independent},
}
\newglossaryentry{ensemble_learning}
{
name=集成学习,
description={ensemble learning},
sort={ensemble learning},
}
\newglossaryentry{NICE}
{
name=非线性独立成分估计,
description={nonlinear independent components estimation},
sort={nonlinear independent components estimation},
symbol={NICE}
}
\newglossaryentry{ISA}
{
name=独立子空间分析,
description={independent subspace analysis},
sort={independent subspace analysis},
}
\newglossaryentry{SFA}
{
name=慢特征分析,
description={slow feature analysis},
sort={slow feature analysis},
symbol={SFA}
}
\newglossaryentry{slow_principle}
{
name=慢性原则,
description={slowness principle},
sort={slowness principle},
}
\newglossaryentry{rectified_linear}
{
name=整流线性,
description={rectified linear},
sort={rectified linear},
}
\newglossaryentry{rectified_linear_transformation}
{
name=整流线性变换,
description={rectified linear transformation},
sort={rectified linear transformation},
}
\newglossaryentry{rectifier_network}
{
name=整流网络,
description={rectifier network},
sort={rectifier network},
}
\newglossaryentry{coordinate_descent}
{
name=坐标下降,
description={coordinate descent},
sort={coordinate descent},
}
\newglossaryentry{coordinate_ascent}
{
name=坐标上升,
description={coordinate ascent},
sort={coordinate ascent},
}
\newglossaryentry{block_coordinate_descent}
{
name=块坐标下降,
description={block coordinate descent},
sort={block coordinate descent},
}
\newglossaryentry{pretraining}
{
name=预训练,
description={pretraining},
sort={pretraining},
}
\newglossaryentry{unsupervised_pretraining}
{
name=无监督预训练,
description={unsupervised pretraining},
sort={unsupervised pretraining},
}
\newglossaryentry{greedy_layer_wise_unsupervised_pretraining}
{
name=贪心逐层无监督预训练,
description={greedy layer-wise unsupervised pretraining},
sort={greedy layer-wise unsupervised pretraining},
}
\newglossaryentry{greedy_layer_wise_pretraining}
{
name=贪心逐层预训练,
description={greedy layer-wise pretraining},
sort={greedy layer-wise pretraining},
}
\newglossaryentry{greedy_layer_wise_training}
{
name=贪心逐层训练,
description={greedy layer-wise training},
sort={greedy layer-wise training},
}
\newglossaryentry{layer_wise}
{
name=逐层的,
description={layer-wise},
sort={layer-wise},
}
\newglossaryentry{greedy_algorithm}
{
name=贪心算法,
description={greedy algorithm},
sort={greedy algorithm},
}
\newglossaryentry{greedy}
{
name=贪心,
description={greedy},
sort={greedy},
}
\newglossaryentry{fine_tuning}
{
name=精调,
description={fine-tuning},
sort={fine-tuning},
}
\newglossaryentry{greedy_supervised_pretraining}
{
name=贪心监督预训练,
description={greedy supervised pretraining},
sort={greedy supervised pretraining},
}
\newglossaryentry{continuation_method}
{
name=延拓法,
description={continuation method},
sort={continuation method},
}
\newglossaryentry{curriculum_learning}
{
name=课程学习,
description={curriculum learning},
sort={curriculum learning},
}
\newglossaryentry{shaping}
{
name=塑造, % ? 整形
description={shaping},
sort={shaping},
}
\newglossaryentry{stochastic_curriculum}
{
name=随机课程,
description={stochastic curriculum},
sort={stochastic curriculum},
}
\newglossaryentry{recall}
{
name=召回率,
description={recall},
sort={recall},
}
\newglossaryentry{coverage}
{
name=覆盖,
description={coverage},
sort={coverage},
}
\newglossaryentry{hyperparameter_optimization}
{
name=超参数优化,
description={hyperparameter optimization},
sort={hyperparameter optimization},
}
\newglossaryentry{hyperparameter}
{
name=超参数,
description={hyperparameter},
sort={hyperparameter}
}
\newglossaryentry{grid_search}
{
name=网格搜索,
description={grid search},
sort={grid search},
}
\newglossaryentry{finite_difference}
{
name=有限差分,
description={finite difference},
sort={finite difference},
}
\newglossaryentry{centered_difference}
{
name=中心差分,
description={centered difference},
sort={centered difference},
}
\newglossaryentry{e_step}
{
name=E步,
description={expectation step},
sort={expectation step},
symbol={E step}
}
\newglossaryentry{m_step}
{
name=M步,
description={maximization step},
sort={maximization step},
symbol={M step}
}
\newglossaryentry{euler_lagrange_eqn}
{
name=欧拉-拉格朗日方程,
description={Euler-Lagrange Equation},
sort={Euler-Lagrange Equation},
}
\newglossaryentry{lagrange_multi}
{
name=拉格朗日乘子,
description={Lagrange multiplier},
sort={Lagrange multiplier},
}
\newglossaryentry{ESN}
{
name=回声状态网络,
description={echo state network},
sort={echo state network},
symbol={ESN}
}
\newglossaryentry{liquid_state_machines}
{
name=流体状态机,
description={liquid state machine},
sort={liquid state machine},
}
\newglossaryentry{reservoir_computing}
{
name=储层计算,
description={reservoir computing},
sort={reservoir computing},
}
\newglossaryentry{spectral_radius}
{
name=谱半径,
description={spectral radius},
sort={spectral radius},
}
\newglossaryentry{contractive}
{
name=收缩,
description={contractive},
sort={contractive},
}
\newglossaryentry{long_term_dependency}
{
name=长期依赖,
description={long-term dependency},
sort={long-term dependency},
}
\newglossaryentry{skip_connection}
{
name=跳跃连接,
description={skip connection},
sort={skip connection},
}
\newglossaryentry{leaky_unit}
{
name=渗漏单元,
description={leaky unit},
sort={leaky unit},
}
\newglossaryentry{gated_rnn}
{
name=门控RNN,
description={gated RNN},
sort={gated RNN},
}
\newglossaryentry{gated}
{
name=门控,
description={gated},
sort={gated},
}
\newglossaryentry{convolution}
{
name=卷积,
description={convolution},
sort={convolution},
}
\newglossaryentry{input}
{
name=输入,
description={input},
sort={input},
}
\newglossaryentry{input_distribution}
{
name=输入分布,
description={input distribution},
sort={input distribution},
}
\newglossaryentry{output}
{
name=输出,
description={output},
sort={output},
}
\newglossaryentry{feature_map}
{
name=特征映射,
description={feature map},
sort={feature map},
}
\newglossaryentry{flip}
{
name=翻转,
description={flip},
sort={flip},
}
\newglossaryentry{cross_correlation}
{
name=互相关函数,
description={cross-correlation},
sort={cross-correlation},
}
\newglossaryentry{Toeplitz_matrix}
{
name=Toeplitz矩阵,
description={Toeplitz matrix},
sort={Toeplitz matrix},
}
\newglossaryentry{doubly_block_circulant_matrix}
{
name=双重分块循环矩阵,
description={doubly block circulant matrix},
sort={doubly block circulant matrix},
}
\newglossaryentry{sparse_interactions}
{
name=稀疏交互,
description={sparse interactions},
sort={sparse interactions},
}
\newglossaryentry{equivariant_representations}
{
name=等变表示,
description={equivariant representations},
sort={equivariant representations},
}
\newglossaryentry{sparse_connectivity}
{
name=稀疏连接,
description={sparse connectivity},
sort={sparse connectivity},
}
\newglossaryentry{sparse_weights}
{
name=稀疏权重,
description={sparse weights},
sort={sparse weights},
}
\newglossaryentry{receptive_field}
{
name=接受域,
description={receptive field},
sort={receptive field},
}
\newglossaryentry{tied_weights}
{
name=绑定的权重,
description={tied weights},
sort={tied weights},
}
\newglossaryentry{equivariance}
{
name=等变,
description={equivariance},
sort={equivariance},
}
\newglossaryentry{detector_stage}
{
name=探测级,
description={detector stage},
sort={detector stage},
}
\newglossaryentry{symbolic_representation}
{
name=符号表示,
description={symbolic representation},
sort={symbolic representation},
}
\newglossaryentry{pooling_funciton}
{
name=池化函数,
description={pooling function},
sort={pooling function},
}
\newglossaryentry{max_pooling}
{
name=最大池化,
description={max pooling},
sort={max pooling},
}
\newglossaryentry{pool}
{
name=池,
description={pool},
sort={max pool},
}
\newglossaryentry{invariant}
{
name=不变,
description={invariant},
sort={invariant},
}
\newglossaryentry{permutation_invariant}
{
name=置换不变性,
description={permutation invariant},
sort={permutation invariant},
}
\newglossaryentry{stride}
{
name=步幅,
description={stride},
sort={stride},
}
\newglossaryentry{downsampling}
{
name=降采样,
description={downsampling},
sort={downsampling},
}
\newglossaryentry{valid}
{
name=有效,
description={valid},
sort={valid},
}
\newglossaryentry{same}
{
name=相同,
description={same},
sort={same},
}
\newglossaryentry{full}
{
name=全,
description={full},
sort={full},
}
\newglossaryentry{unshared_convolution}
{
name=非共享卷积,
description={unshared convolution},
sort={unshared convolution},
}
\newglossaryentry{tiled_convolution}
{
name=平铺卷积,
description={tiled convolution},
sort={tiled convolution},
}
\newglossaryentry{recurrent_convolutional_network}
{
name=循环卷积网络,
description={recurrent convolutional network},
sort={recurrent convolutional network},
}
\newglossaryentry{Fourier_transform}
{
name=傅立叶变换,
description={Fourier transform},
sort={Fourier transform},
}
\newglossaryentry{separable}
{
name=可分离的,
description={separable},
sort={separable},
}
\newglossaryentry{primary_visual_cortex}
{
name=初级视觉皮层,
description={primary visual cortex},
sort={primary visual cortex},
}
\newglossaryentry{simple_cells}
{
name=简单细胞,
description={simple cell},
sort={simple cell},
}
\newglossaryentry{complex_cells}
{
name=复杂细胞,
description={complex cell},
sort={complex cell},
}
\newglossaryentry{fovea}
{
name=中央凹,
description={fovea},
sort={fovea},
}
\newglossaryentry{saccade}
{
name=扫视,
description={saccade},
sort={saccade},
}
\newglossaryentry{TDNNs}
{
name=时延神经网络,
description={time delay neural network},
sort={time delay neural network},
symbol={TDNN}
}
\newglossaryentry{reverse_correlation}
{
name=反向相关,
description={reverse correlation},
sort={reverse correlation},
}
\newglossaryentry{Gabor_function}
{
name=Gabor函数,
description={Gabor function},
sort={Gabor function},
}
\newglossaryentry{quadrature_pair}
{
name=象限对,
description={quadrature pair},
sort={quadrature pair},
}
\newglossaryentry{gated_recurrent_unit}
{
name=门控循环单元,
description={gated recurrent unit},
sort={gated recurrent unit},
symbol={GRU}
}
\newglossaryentry{gated_recurrent_net}
{
name=门控循环网络,
description={gated recurrent net},
sort={gated recurrent net},
}
\newglossaryentry{forget_gate}
{
name=遗忘门,
description={forget gate},
sort={forget gate},
}
\newglossaryentry{clipping_gradient}
{
name=截断梯度,
description={clipping the gradient},
sort={clipping the gradient},
}
\newglossaryentry{memory_network}
{
name=记忆网络,
description={memory network},
sort={memory network},
}
\newglossaryentry{NTM}
{
name=神经网络图灵机,
description={neural Turing machine},
sort={neural Turing machine},
symbol={NTM}
}
\newglossaryentry{fine_tune}
{
name=精调,
description={fine-tune},
sort={fine-tune},
}
\newglossaryentry{explaining_away}
{
name=相消解释,
description={explaining away},
sort={explaining away},
}
\newglossaryentry{explaining_away_effect}
{
name=相消解释作用,
description={explaining away effect},
sort={explaining away effect},
}
\newglossaryentry{common_cause}
{
name=共因,
description={common cause},
sort={common cause},
}
\newglossaryentry{code}
{
name=编码,
description={code},
sort={code},
}
\newglossaryentry{recirculation}
{
name=再循环,
description={recirculation},
sort={recirculation},
}
\newglossaryentry{undercomplete}
{
name=欠完备,
description={undercomplete},
sort={undercomplete},
}
\newglossaryentry{complete_graph}
{
name=完全图,
description={complete graph},
sort={complete graph},
}
\newglossaryentry{underdetermined}
{
name=欠定的,
description={underdetermined},
sort={underdetermined},
}
\newglossaryentry{overcomplete}
{
name=过完备,
description={overcomplete},
sort={overcomplete},
}
\newglossaryentry{denoising}
{
name=去噪,
description={denoising},
sort={denoising},
}
\newglossaryentry{denoise}
{
name=去噪,
description={denoise}
}
\newglossaryentry{DAE}
{
name=去噪自编码器,
description={denoising autoencoder},
sort={denoising autoencoder},
symbol={DAE}
}
\newglossaryentry{CAE}
{
name=收缩自编码器,
description={contractive autoencoder},
sort={contractive autoencoder},
symbol={CAE}
}
\newglossaryentry{reconstruction_error}
{
name=重构误差,
description={reconstruction error},
sort={reconstruction error},
}
\newglossaryentry{gradient_field}
{
name=梯度场,
description={gradient field},
sort={gradient field},
}
\newglossaryentry{denoising_score_matching}
{
name=去噪得分匹配,
description={denoising score matching},
sort={denoising score matching},
}
\newglossaryentry{score}
{
name=得分,
description={score},
sort={score},
}
\newglossaryentry{tangent_plane}
{
name=切平面,
description={tangent plane},
sort={tangent plane},
}
\newglossaryentry{nearest_neighbor_graph}
{
name=最近邻图,
description={nearest neighbor graph},
sort={nearest neighbor graph},
}
\newglossaryentry{embedding}
{
name=嵌入,
description={embedding},
sort={embedding},
}
\newglossaryentry{PSD}
{
name=预测稀疏分解,
description={predictive sparse decomposition},
sort={predictive sparse decomposition},
symbol={PSD}
}
\newglossaryentry{learned_approximate_inference}
{
name=学习近似推断,
description={learned approximate inference},
sort={learned approximate inference},
}
\newglossaryentry{approximate_inference}
{
name=近似推断,
description={approximate inference},
sort={approximate inference},
}
\newglossaryentry{information_retrieval}
{
name=信息检索,
description={information retrieval},
sort={information retrieval},
}
\newglossaryentry{semantic_hashing}
{
name=语义哈希,
description={semantic hashing},
sort={semantic hashing},
}
\newglossaryentry{dimensionality_reduction}
{
name=降维,
description={dimensionality reduction},
sort={dimensionality reduction},
}
\newglossaryentry{helmholtz_machine}
{
name=Helmholtz机,
description={Helmholtz machine},
sort={Helmholtz machine},
}
\newglossaryentry{contrastive_divergence}
{
name=对比散度,
description={contrastive divergence},
symbol={CD},
sort={contrastive divergence},
}
\newglossaryentry{language_model}
{
name=语言模型,
description={language model},
sort={language model},
}
\newglossaryentry{token}
{
name=标记,
description={token},
sort={token},
}
\newglossaryentry{unigram}
{
name=一元语法,
description={unigram},
sort={unigram},
}
\newglossaryentry{bigram}
{
name=二元语法,
description={bigram},
sort={bigram},
}
\newglossaryentry{trigram}
{
name=三元语法,
description={trigram},
sort={trigram},
}
\newglossaryentry{smoothing}
{
name=平滑,
description={smoothing},
sort={smoothing},
}
\newglossaryentry{NLM}
{
name=神经语言模型,
description={Neural Language Model},
sort={Neural Language Model},
symbol={NLM}
}
\newglossaryentry{cascade}
{
name=级联,
description={cascade},
sort={cascade},
}
\newglossaryentry{policy_gradient}
{
name=策略梯度,
description={policy gradient},
sort={policy gradient},
}
\newglossaryentry{DGM}
{
name=深度生成模型,
description={deep generative model},
sort={deep generative model},
}
\newglossaryentry{model}
{
name=模型,
description={model},
sort={model},
}
\newglossaryentry{layer}
{
name=层,
description={layer},
sort={layer},
}
\newglossaryentry{greedy_unsupervised_pretraining}
{
name=贪心无监督预训练,
description={greedy unsupervised pretraining},
sort={greedy unsupervised pretraining},
}
\newglossaryentry{semi_supervised_learning}
{
name=半监督学习,
description={semi-supervised learning},
sort={semi-supervised learning},
}
\newglossaryentry{supervised_model}
{
name=监督模型,
description={supervised model},
sort={supervised model},
}
\newglossaryentry{word_embeddings}
{
name=词嵌入,
description={word embedding},
sort={word embedding},
}
\newglossaryentry{one_hot}
{
name=one-hot,
description={one-hot},
sort={one-hot},
}
\newglossaryentry{supervised_pretraining}
{
name=监督预训练,
description={supervised pretraining},