add resources

_posts/artificial_intelligence/2015-10-09-ai-resources.md

@@ -23,6 +23,10 @@ date: 2015-10-01
 
 - homepage: [https://www.edx.org/course/artificial-intelligence-ai-columbiax-csmm-101x](https://www.edx.org/course/artificial-intelligence-ai-columbiax-csmm-101x)
 
+**MIT 6.S099: Artificial General Intelligence**
+
+[https://agi.mit.edu/](https://agi.mit.edu/)
+
 # Books
 
 **Notes on Artificial Intelligence (open source notebook)**
@@ -134,4 +138,4 @@ We aim to inspire a new generation of research into challenging new problems pre
 **WHAT-AI-CAN-DO-FOR-YOU**
 
 - intro: Breakthrough AI Papers and CODE for Any Industry.
-- github: [https://github.com/ceobillionaire/WHAT-AI-CAN-DO-FOR-YOU](https://github.com/ceobillionaire/WHAT-AI-CAN-DO-FOR-YOU)
+- github: [https://github.com/ceobillionaire/WHAT-AI-CAN-DO-FOR-YOU](https://github.com/ceobillionaire/WHAT-AI-CAN-DO-FOR-YOU)

_posts/deep_learning/2015-10-09-acceleration-model-compression.md

@@ -447,6 +447,12 @@ at INT8 deep learning operations than other FPGA DSP architectures"
 
 [https://arxiv.org/abs/1711.06528](https://arxiv.org/abs/1711.06528)
 
+**Reducing Deep Network Complexity with Fourier Transform Methods**
+
+- intro: Harvard University
+- arxiv: [https://arxiv.org/abs/1801.01451](https://arxiv.org/abs/1801.01451)
+- github: [https://github.com/andrew-jeremy/Reducing-Deep-Network-Complexity-with-Fourier-Transform-Methods](https://github.com/andrew-jeremy/Reducing-Deep-Network-Complexity-with-Fourier-Transform-Methods)
+
 # Pruning
 
 **ThiNet: A Filter Level Pruning Method for Deep Neural Network Compression**
@@ -603,6 +609,11 @@ at INT8 deep learning operations than other FPGA DSP architectures"
 - intro: AAAI 2018
 - arxiv: [https://arxiv.org/abs/1712.07493](https://arxiv.org/abs/1712.07493)
 
+**SBNet: Sparse Blocks Network for Fast Inference**
+
+- intro: Uber
+- arxiv: [https://arxiv.org/abs/1801.02108](https://arxiv.org/abs/1801.02108)
+
 # Knowledge Distilling / Knowledge Transfer
 
 **Distilling the Knowledge in a Neural Network**

_posts/deep_learning/2015-10-09-autonomous-driving.md

@@ -211,6 +211,11 @@ without running into obstacles using a type of reinforcement learning called Q-L
 - arxiv: [https://arxiv.org/abs/1712.08644](https://arxiv.org/abs/1712.08644)
 - github: [https://github.com//heechul/picar](https://github.com//heechul/picar)
 
+**Autonomous Driving in Reality with Reinforcement Learning and Image Translation**
+
+- intro: Shanghai Jiao Tong University
+- arxiv: [https://arxiv.org/abs/1801.05299](https://arxiv.org/abs/1801.05299)
+
 # Blogs
 
 **Self-driving cars: How far away are we REALLY from autonomous cars?(7 Aug 2015)**

_posts/deep_learning/2015-10-09-dl-applications.md

@@ -455,6 +455,11 @@ Illustration Tagger, InceptionNet, Parsey McParseface, ArtsyNetworks
 - arxiv: [https://arxiv.org/abs/1708.01101](https://arxiv.org/abs/1708.01101)
 - github: [https://github.com/bearpaw/PyraNet](https://github.com/bearpaw/PyraNet)
 
+**Crossing Nets: Combining GANs and VAEs with a Shared Latent Space for Hand Pose Estimation**
+
+- intro: CVPR 2017
+- arxiv: [https://arxiv.org/abs/1702.03431](https://arxiv.org/abs/1702.03431)
+
 **Multi-Context Attention for Human Pose Estimation**
 
 - intro: CVPR 2017
@@ -1229,6 +1234,10 @@ Illustration Tagger, InceptionNet, Parsey McParseface, ArtsyNetworks
 - intro: GCPR 2017
 - arxiv: [https://arxiv.org/abs/1707.00471](https://arxiv.org/abs/1707.00471)
 
+**Frame-Recurrent Video Super-Resolution**
+
+[https://arxiv.org/abs/1801.04590](https://arxiv.org/abs/1801.04590)
+
 # Image Denoising
 
 **Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising**
@@ -1347,6 +1356,10 @@ Illustration Tagger, InceptionNet, Parsey McParseface, ArtsyNetworks
 
 [https://arxiv.org/abs/1711.06787](https://arxiv.org/abs/1711.06787)
 
+**CANDY: Conditional Adversarial Networks based Fully End-to-End System for Single Image Haze Removal**
+
+[https://arxiv.org/abs/1801.02892](https://arxiv.org/abs/1801.02892)
+
 # Image Rain Removal / De-raining
 
 **Clearing the Skies: A deep network architecture for single-image rain removal**
@@ -1439,6 +1452,11 @@ Illustration Tagger, InceptionNet, Parsey McParseface, ArtsyNetworks
 
 [https://arxiv.org/abs/1711.09515](https://arxiv.org/abs/1711.09515)
 
+**Reblur2Deblur: Deblurring Videos via Self-Supervised Learning**
+
+- arxiv: [https://arxiv.org/abs/1801.05117](https://arxiv.org/abs/1801.05117)
+- supplementary: [https://drive.google.com/file/d/17Itta-z89lpWUdvUjpafKzJRSLoxHF5c/view](https://drive.google.com/file/d/17Itta-z89lpWUdvUjpafKzJRSLoxHF5c/view)
+
 # Image Compression
 
 **An image compression and encryption scheme based on deep learning**
@@ -1495,6 +1513,10 @@ Illustration Tagger, InceptionNet, Parsey McParseface, ArtsyNetworks
 - project page: [http://www.wave.one/icml2017](http://www.wave.one/icml2017)
 - arxiv: [https://arxiv.org/abs/1705.05823](https://arxiv.org/abs/1705.05823)
 
+**Efficient Trimmed Convolutional Arithmetic Encoding for Lossless Image Compression**
+
+[https://arxiv.org/abs/1801.04662](https://arxiv.org/abs/1801.04662)
+
 # Image Quality Assessment
 
 **Deep Neural Networks for No-Reference and Full-Reference Image Quality Assessment**
@@ -1596,6 +1618,10 @@ Illustration Tagger, InceptionNet, Parsey McParseface, ArtsyNetworks
 - video: [https://www.youtube.com/watch?v=vNIIT_M7x7Y](https://www.youtube.com/watch?v=vNIIT_M7x7Y)
 - github: [https://github.com/fangchangma/sparse-to-dense](https://github.com/fangchangma/sparse-to-dense)
 
+**Size-to-depth: A New Perspective for Single Image Depth Estimation**
+
+[https://arxiv.org/abs/1801.04461](https://arxiv.org/abs/1801.04461)
+
 # Texture Synthesis
 
 **Texture Synthesis Using Convolutional Neural Networks**
@@ -2138,39 +2164,6 @@ Illustration Tagger, InceptionNet, Parsey McParseface, ArtsyNetworks
 
 [https://arxiv.org/html/1706.08675](https://arxiv.org/html/1706.08675)
 
-# Deep Learning on Games
-
-**TorchCraft: a Library for Machine Learning Research on Real-Time Strategy Games**
-
-- intro: Connecting Torch to StarCraft
-- arxiv: [https://arxiv.org/abs/1611.00625](https://arxiv.org/abs/1611.00625)
-- github: [https://github.com/TorchCraft/TorchCraft](https://github.com/TorchCraft/TorchCraft)
-
-**BlizzCon 2016 DeepMind and StarCraft II Deep Learning Panel Transcript**
-
-- part 1: [http://starcraft.blizzplanet.com/blog/comments/blizzcon-2016-deepmind-and-starcraft-ii-deep-learning-panel-transcript](http://starcraft.blizzplanet.com/blog/comments/blizzcon-2016-deepmind-and-starcraft-ii-deep-learning-panel-transcript)
-- part 2: [http://starcraft.blizzplanet.com/blog/comments/blizzcon-2016-deepmind-and-starcraft-ii-deep-learning-panel-transcript/2](http://starcraft.blizzplanet.com/blog/comments/blizzcon-2016-deepmind-and-starcraft-ii-deep-learning-panel-transcript/2)
-
-**DeepStack: Expert-Level Artificial Intelligence in No-Limit Poker**
-
-- arxiv: [https://arxiv.org/abs/1701.01724](https://arxiv.org/abs/1701.01724)
-- github: [https://github.com/lifrordi/DeepStack-Leduc](https://github.com/lifrordi/DeepStack-Leduc)
-
-**Gym StarCraft: StarCraft environment for OpenAI Gym, based on Facebook's TorchCraft**
-
-- intro: Gym StarCraft is an environment bundle for OpenAI Gym. 
-It is based on Facebook's TorchCraft, which is a bridge between Torch and StarCraft for AI research.
-- github: [https://github.com/deepcraft/gym-starcraft](https://github.com/deepcraft/gym-starcraft)
-
-**Multiagent Bidirectionally-Coordinated Nets for Learning to Play StarCraft Combat Games**
-
-[https://arxiv.org/abs/1703.10069](https://arxiv.org/abs/1703.10069)
-
-**Learning Macromanagement in StarCraft from Replays using Deep Learning**
-
-- intro: CIG 2017. IT University of Copenhagen
-- arxiv: [https://arxiv.org/abs/1707.03743](https://arxiv.org/abs/1707.03743)
-
 # Deep Learning in Medicine and Biology
 
 **Low Data Drug Discovery with One-shot Learning**

_posts/deep_learning/2015-10-09-dl-courses.md

@@ -110,7 +110,7 @@ date: 2015-10-09
 **CS 20SI: Tensorflow for Deep Learning Research**
 
 - homepage: [http://web.stanford.edu/class/cs20si/](http://web.stanford.edu/class/cs20si/)
-- github: [https://github.com/chiphuyen/tf-stanford-tutorials](https://github.com/chiphuyen/tf-stanford-tutorials)
+- github: [https://github.com/chiphuyen/stanford-tensorflow-tutorials](https://github.com/chiphuyen/stanford-tensorflow-tutorials)
 
 **Deep Learning with TensorFlow**
 

_posts/deep_learning/2015-10-09-dl-frameworks.md

@@ -466,6 +466,11 @@ It is an experimental refactoring of Caffe, and allows a more flexible way to or
 
 - github: [https://github.com/pppoe/WhatsThis-iOS](https://github.com/pppoe/WhatsThis-iOS)
 
+**MXNET-MPI: Embedding MPI parallelism in Parameter Server Task Model for scaling Deep Learning**
+
+- intro: IBM T J Watson Research Center
+- arxiv: [https://arxiv.org/abs/1801.03855](https://arxiv.org/abs/1801.03855)
+
 # ncnn
 
 - intro: ncnn is a high-performance neural network inference framework optimized for the mobile platform

_posts/deep_learning/2015-10-09-dl-resources.md

@@ -341,6 +341,8 @@ not just the convolutions)
 - github: [https://github.com/taineleau/efficient_densenet_mxnet](https://github.com/taineleau/efficient_densenet_mxnet)
 - github: [https://github.com/Tongcheng/DN_CaffeScript](https://github.com/Tongcheng/DN_CaffeScript)
 
+## DenseNet 2.0
+
 **CondenseNet: An Efficient DenseNet using Learned Group Convolutions**
 
 - arxiv: [https://arxiv.org/abs/1711.09224](https://arxiv.org/abs/1711.09224)
@@ -350,10 +352,10 @@ not just the convolutions)
 
 **Squeeze-and-Excitation Networks**
 
-- intro: ILSVRC 2017 image classification winner
+- intro: ILSVRC 2017 image classification winner. Momenta & University of Oxford
 - arxiv: [https://arxiv.org/abs/1709.01507](https://arxiv.org/abs/1709.01507)
-- github: [https://github.com/hujie-frank/SENet](https://github.com/hujie-frank/SENet)
-- github: [https://github.com//bruinxiong/SENet.mxnet](https://github.com//bruinxiong/SENet.mxnet)
+- github(official, Caffe): [https://github.com/hujie-frank/SENet](https://github.com/hujie-frank/SENet)
+- github: [https://github.com/bruinxiong/SENet.mxnet](https://github.com/bruinxiong/SENet.mxnet)
 
 ## ImageNet Projects
 
@@ -938,6 +940,25 @@ not just the convolutions)
 
 [https://arxiv.org/abs/1801.00553](https://arxiv.org/abs/1801.00553)
 
+**Spatially transformed adversarial examples**
+
+[https://arxiv.org/abs/1801.02612](https://arxiv.org/abs/1801.02612)
+
+**Generating adversarial examples with adversarial networks**
+
+- intro: University of Michigan & UC Berkeley & MIT CSAIL
+- arxiv: [https://arxiv.org/abs/1801.02610](https://arxiv.org/abs/1801.02610)
+
+**Adversarial Spheres**
+
+- intro: Google Brain
+- arxiv: [https://arxiv.org/abs/1801.02774](https://arxiv.org/abs/1801.02774)
+
+**LaVAN: Localized and Visible Adversarial Noise**
+
+- intro: Bar-Ilan University & DeepMind
+- arxiv: [https://arxiv.org/abs/1801.02608](https://arxiv.org/abs/1801.02608)
+
 # Deep Learning Networks
 
 **PCANet: A Simple Deep Learning Baseline for Image Classification?**
@@ -1413,6 +1434,10 @@ with fast exact probabilistic inference over many layers."
 - intro: NIPS 2017 Symposium on Interpretable Machine Learning
 - arxiv: [https://arxiv.org/abs/1711.02329](https://arxiv.org/abs/1711.02329)
 
+**Interpreting Deep Neural Networks**
+
+- blog: [http://www.shallowmind.co/jekyll/pixyll/2017/12/30/tree-regularization/](http://www.shallowmind.co/jekyll/pixyll/2017/12/30/tree-regularization/)
+
 ## Convolutions / Filters
 
 **Warped Convolutions: Efficient Invariance to Spatial Transformations**
@@ -1433,6 +1458,11 @@ with fast exact probabilistic inference over many layers."
 
 [https://arxiv.org/abs/1712.06145](https://arxiv.org/abs/1712.06145)
 
+**Non-Parametric Transformation Networks**
+
+- intro: CMU
+- arxiv: [https://arxiv.org/abs/1801.04520](https://arxiv.org/abs/1801.04520)
+
 ## Highway Networks
 
 **Highway Networks**
@@ -1682,6 +1712,16 @@ with fast exact probabilistic inference over many layers."
 - mirror: [https://www.bilibili.com/video/av16428277/](https://www.bilibili.com/video/av16428277/)
 - slides: [http://www.shakirm.com/slides/DeepGenModelsTutorial.pdf](http://www.shakirm.com/slides/DeepGenModelsTutorial.pdf)
 
+**A Note on the Inception Score**
+
+- intro: Stanford University
+- arxiv: [https://arxiv.org/abs/1801.01973](https://arxiv.org/abs/1801.01973)
+
+**Gradient Layer: Enhancing the Convergence of Adversarial Training for Generative Models**
+
+- intro: AISTATS 2018. The University of Tokyo
+- arxiv: [https://arxiv.org/abs/1801.02227](https://arxiv.org/abs/1801.02227)
+
 # Deep Learning and Robots
 
 **Robot Learning Manipulation Action Plans by "Watching" Unconstrained Videos from the World Wide Web**
@@ -2278,6 +2318,17 @@ with fast exact probabilistic inference over many layers."
 
 [https://openreview.net/forum?id=ry_WPG-A-&noteId=ry_WPG-A](https://openreview.net/forum?id=ry_WPG-A-&noteId=ry_WPG-A)
 
+**The Unreasonable Effectiveness of Deep Features as a Perceptual Metric**
+
+- project page: [https://richzhang.github.io/PerceptualSimilarity/](https://richzhang.github.io/PerceptualSimilarity/)
+- arxiv: [https://arxiv.org/abs/1801.03924](https://arxiv.org/abs/1801.03924)
+- github: [https://github.com//richzhang/PerceptualSimilarity](https://github.com//richzhang/PerceptualSimilarity)
+
+**Less is More: Culling the Training Set to Improve Robustness of Deep Neural Networks**
+
+- intro: University of California, Davis
+- arxiv: [https://arxiv.org/abs/1801.02850](https://arxiv.org/abs/1801.02850)
+
 ## Tutorials and Surveys
 
 **On the Origin of Deep Learning**
@@ -2394,6 +2445,14 @@ with fast exact probabilistic inference over many layers."
 
 [https://arxiv.org/abs/1712.04698](https://arxiv.org/abs/1712.04698)
 
+## MobileNetV2
+
+**Inverted Residuals and Linear Bottlenecks: Mobile Networks forClassification, Detection and Segmentation**
+
+- intro: Google
+- keywords: MobileNetV2, SSDLite, DeepLabv3
+- arxiv: [https://arxiv.org/abs/1801.04381](https://arxiv.org/abs/1801.04381)
+
 ## STDP
 
 **A biological gradient descent for prediction through a combination of STDP and homeostatic plasticity**
@@ -2635,6 +2694,14 @@ joint classification, detection and semantic segmentation via a unified architec
 - arxiv: [https://arxiv.org/abs/1612.07695](https://arxiv.org/abs/1612.07695)
 - github: [https://github.com/MarvinTeichmann/MultiNet](https://github.com/MarvinTeichmann/MultiNet)
 
+### Deep Learning for Data Structures
+
+**The Case for Learned Index Structures**
+
+- intro: MIT & Google
+- keywords: B-Tree-Index, Hash-Index, BitMap-Index
+- arxiv: [https://arxiv.org/abs/1712.01208](https://arxiv.org/abs/1712.01208)
+
 # Projects
 
 **Top Deep Learning Projects**

_posts/deep_learning/2015-10-09-dl-software-hardware.md

@@ -296,6 +296,11 @@ Contains all the popular DL frameworks (TensorFlow, Theano, Torch, Caffe, etc.)
 - intro: University of Birmingham
 - arxiv: [https://arxiv.org/abs/1711.05860](https://arxiv.org/abs/1711.05860)
 
+**Approximate FPGA-based LSTMs under Computation Time Constraints**
+
+- intro: ARC 2018
+- arxiv: [https://arxiv.org/abs/1801.02190](https://arxiv.org/abs/1801.02190)
+
 # ARM / Processor
 
 **'Neural network' spotted deep inside Samsung's Galaxy S7 silicon brain: Secrets of Exynos M1 cores spilled**

_posts/deep_learning/2015-10-09-dl-with-ml.md

@@ -155,6 +155,10 @@ date: 2015-10-09
 
 - arxiv: [https://arxiv.org/abs/1702.02549](https://arxiv.org/abs/1702.02549)
 
+**Semi-supervised Fisher vector network**
+
+[https://arxiv.org/abs/1801.04438](https://arxiv.org/abs/1801.04438)
+
 # Gaussian Processes
 
 **Questions on Deep Gaussian Processes**
@@ -194,6 +198,11 @@ date: 2015-10-09
 - intro: NIPS workshop on Advances in Approximate Bayesian Inference 2017
 - slide: [http://adamian.github.io/talks/Damianou_NIPS17.pdf](http://adamian.github.io/talks/Damianou_NIPS17.pdf)
 
+**Deep Gaussian Processes with Decoupled Inducing Inputs**
+
+- intro: University of Cambridge & University of Seville
+- arxiv: [https://arxiv.org/abs/1801.02939](https://arxiv.org/abs/1801.02939)
+
 ## Graphical Models
 
 **GibbsNet: Iterative Adversarial Inference for Deep Graphical Models**