Deep-Learning-Papers [Incomplete List]

This repository consist of various Deep Learning papers sorted by subareas. Currently this list is incomplete , I will add them overtime. I will also update this repository as I come across more papers that are interesting. Please feel free to make a pull request and add to the list.

General Deep Learning

• Patterns, predictions, and actions: A story about machine learning
• On Calibration of Modern Neural Networks
• Understanding deep learning requires rethinking generalization
• Label Refinery: Improving ImageNet Classification through Label Progression

Industrial Machine Learning

• A Github Repository Repository that did a great job curating resources

Generalization

• Do CIFAR-10 Classifiers Generalize to CIFAR-10?

Neural Architecture Search

• BANANAS: Bayesian Optimization with Neural Architectures for Neural Architecture Search

Explainability

• Explainable Machine Learning in Deployment

Natural Language Processing

GPT

• GPT-1: Improving Language Understanding by Generative Pre-Training
• GPT-2: Language Models are Unsupervised Multitask Learners
• GPT-3: Language Models are Few-Shot Learners

Computer Vision

Imagenet Architectures

• Alexnet: ImageNet Classification with Deep Convolutional Neural Networks
• VGG: Very Deep Convolutional Networks for Large-Scale Image Recognition
• Resnet: Deep Residual Learning for Image Recognition
• Densenet: Densely Connected Convolutional Networks
• Mobilenet: MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Squeezenet: SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size

Scene Understanding

• SUN RGB-D: A RGB-D Scene Understanding Benchmark Suite

Weakly Supervised Object Detection

• Multiple Instance Detection Network with Online Instance Classifier Refinement
• Weakly Supervised Object Detection with Segmentation Collaboration
• Weakly Supervised Deep Detection Networks

Understanding and Visualizing

• Visualizing and Understanding Convolutional Networks
• Do Wide and Deep Networks Learn the Same Things? Uncovering How Neural Network Representations Vary with Width and Depth

Faster Training

• LARGE BATCH TRAINING OF CONVOLUTIONAL NETWORKS : Introduces adaptive layerwise learning rates. Different learning rates for different layers.
• Accelerating Deep Learning by Focusing on the Biggest Losers

Meta Learning

• MAML: Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Reptile: On First-Order Meta-Learning Algorithms

Multitask Learning

• Recommending What Video to Watch Next: A Multitask Ranking System
• A Survey on Multi-Task Learning
• Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics

Zero shot Learning

• Zero-Shot Learning -- The Good, the Bad and the Ugly
• Transductive Unbiased Embedding for Zero-Shot Learning
• Semantic Autoencoder for Zero-Shot Learning

Bayesian Deep Learning

• Bayesian Convolutional Neural Networks with Bernoulli Approximate Variational Inference
• What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?

Privacy Preserving Learning

Federated Learning: Data Security

• Communication-Efficient Learning of Deep Networks from Decentralized Data
• Federated Learning: Strategies for Improving Communication Efficiency
• Practical Secure Aggregation for Privacy-Preserving Machine Learning
• Active Federated Learning
• Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training
• Federated Learning with Non-IID Data
• Advances and Open Problems in Federated Learning
• Addressing Class Imbalance in Federated Learning

Model Security (Inference)

• DeepObfuscation: Securing the Structure of Convolutional Neural Networks via Knowledge Distillation

Privacy Attacks & Defenses

• Evaluating and Testing Unintended Memorization in Neural Networks
• Privacy Considerations in Large Language Models

Neural Network Compression

• A Survey of Model Compression and Acceleration for Deep Neural Networks
• Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding
• The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks
• Distilling the Knowledge in a Neural Network
• FitNets: Hints for Thin Deep Nets