refactor: split content 'code' and 'papers'

Author: DongjunLee

SUMMARY.md

@@ -1,6 +1,9 @@
 # Table of contents
 
 * [What is notes](README.md)
+
+## Basic
+
 * [Math]()
     * [Statictics]()
     * [Linear Algebra]()
@@ -9,22 +12,21 @@
 
 ## Code
 
-* [Computer Vision]()
-* [Natural Language Processing]()
-* [Reinforcement Learning]()
-* [Unsupervised & Generative]()
+* [Generative](code/generative.md)
+* [NLP](code/nlp.md)
+* [RL](code/rl.md)
+* [Vision](code/vision.md)
 
 ## Research Paper
 
-* [Adversarial Example]()
-* [AI]()
-* [Cognitive]()
-* [Computer Vision]()
-* [Framework & System]()
-* [Model]()
-* [Natural Language Processing]()
-* [One-Shot/Few-Shot Learing]()
-* [Optimization]()
-* [Reinforcement Learning]()
-* [Transfer Learning]()
-* [Unsupervised & Generative]()
+* [Adversarial](papers/Adversarial.md)
+* [AI](papers/AI.md)
+* [Cognitive](papers/Cognitive.md)
+* [Computer Vision](papers/Computer_Vision.md)
+* [Framework & System](papers/Framework_System.md)
+* [Model](papers/Model.md)
+* [Natural Language Processing](papers/Natural_Language_Processing.md)
+* [One-Shot/Few-Shot Learing](papers/One-Shot_Few-Shot_Learing.md)
+* [Optimization](papers/Optimization.md)
+* [Reinforcement Learning](papers/Reinforcement_Learning.md)
+* [Unsupervised & Generative](papers/Unsupervised_Generative.md)

code/generative.md

@@ -0,0 +1,7 @@
+### PyTorch
+
+- [gan-pytorch](https://github.com/DongjunLee/gan-pytorch) : Generative Adversarial Networks
+
+### TensorFlow
+
+- [vae-tensorflow](https://github.com/DongjunLee/vae-tensorflow) : Auto-Encoding Variational Bayes

code/nlp.md

@@ -0,0 +1,10 @@
+### TensorFlow
+
+The codes are implemented by TensorFlow and initiate project with [hb-base](https://github.com/DongjunLee/hb-base).
+
+- [transformer-tensorflow](https://github.com/DongjunLee/transformer-tensorflow) : Attention Is All You Need
+- [relation-network-tensorflow](https://github.com/DongjunLee/relation-network-tensorflow) : A simple neural network module for relational reasoning for bAbi task
+- [conversation-tensorflow](https://github.com/DongjunLee/conversation-tensorflow) : Conversation Models (Seq2Seq with Attentional Model)
+- [dmn-tensorflow](https://github.com/DongjunLee/dmn-tensorflow) : Ask Me Anything: Dynamic Memory Networks for Natural Language Processing
+- [text-cnn-tensorflow](https://github.com/DongjunLee/text-cnn-tensorflow) : Convolutional Neural Networks for Sentence Classification(TextCNN)
+- [char-rnn-tensorflow](https://github.com/DongjunLee/char-rnn-tensorflow) : Multi-layer Recurrent Neural Networks for character-level language models

code/rl.md

@@ -0,0 +1,3 @@
+### TensorFlow
+
+- [dqn-tensorflow](https://github.com/DongjunLee/dqn-tensorflow) : Playing Atari with Deep Reinforcement Learning (2013) & Human-Level Control through Deep Reinforcement Learning (2015)

code/vision.md

@@ -0,0 +1,5 @@
+### Papers
+
+- Machine Theory of Mind (2018. 2)
+	- **`ToMnet`**, **`Meta-Learning`**, **`General Model`**, **`Agent`**
+	- [arXiv](https://arxiv.org/abs/1802.07740)

papers/AI.md

@@ -0,0 +1,17 @@
+### Papers
+
+- Explaining and Harnessing Adversarial Examples (2014. 12)
+	- **`FGSM (Fast Gradient Sign Method)`**, **`Adversarial Training`**
+	- [arXiv](https://arxiv.org/abs/1412.6572)
+- The Limitations of Deep Learning in Adversarial Settings (2015. 11)
+	- **`JSMA (Jacobian-based Saliency Map Approach)`**, **`Adversarial Training`**
+	- [arXiv](https://arxiv.org/abs/1511.07528)
+- Understanding Adversarial Training: Increasing Local Stability of Neural Nets through Robust Optimization (2015. 11)
+	- **`Adversarial Training (generated adversarial examples)`**, **`Proactive Defense`**
+	- [arXiv](https://arxiv.org/abs/1511.05432)
+- Practical Black-Box Attacks against Machine Learning (2016. 2)
+	- **`Black-Box (No Access to Gradient)`**, **`Generate Synthetic`**
+	- [arXiv](https://arxiv.org/abs/1602.02697)
+- Adversarial Patch (2017. 12)
+	- **`Patch`**, **`White Box`**, **`Black Box`**
+	- [arXiv](https://arxiv.org/abs/1712.09665), [the_morning_paper](https://blog.acolyer.org/2018/03/29/adversarial-patch/)

papers/Adversarial.md

@@ -0,0 +1,5 @@
+### Papers
+
+- Building Machines That Learn and Think Like People (2016. 4)
+	- **`Human-Like`**, **`Learn`**, **`Think`**
+	- [arXiv](https://arxiv.org/abs/1604.00289), [note](notes/ml_learn_and_think_like_human.md), [the morning paper](https://blog.acolyer.org/2016/11/25/building-machines-that-learn-and-think-like-people/)

papers/Cognitive.md

@@ -0,0 +1,20 @@
+### Papers
+
+- **Network In Network** (2013. 12)
+	- **`Conv 1x1`**, **`Bottleneck`**
+	- [arXiv](https://arxiv.org/abs/1312.4400), [note](notes/network_in_network.md)
+- Fractional Max-Pooling (2014. 12)
+	- **`Max-Pooling`**, **`Data Augmentation`**, **`Regularization`**
+	- [arXiv](https://arxiv.org/abs/1412.6071), [note](notes/fractional_max-pooling.md)
+- **Deep Residual Learning for Image Recognition** (2015. 12)
+	- **`Residual`**, **`ImageNet 2015`**
+	- [arXiv](https://arxiv.org/abs/1512.03385), [note](notes/residual_network.md)
+- Spherical CNNs (2018. 1)
+	- **`Spherical Correlation`**, **`3D Model`**, **`Fast Fourier Transform (FFT)`**
+	- [arXiv](https://arxiv.org/abs/1801.10130), [open_review](https://openreview.net/forum?id=Hkbd5xZRb)
+- Taskonomy: Disentangling Task Transfer Learning (2018. 4)
+	- **`Taskonomy `**, **`Transfer Learning`**, **`Computational modeling of task relations`**
+	- [arXiv](https://arxiv.org/abs/1804.08328) 
+- AutoAugment: Learning Augmentation Policies from Data (2018. 5)
+	- **`Search Algorithm (RL)`**, **`Sub-Policy`**
+	- [arXiv](https://arxiv.org/abs/1805.09501)

papers/Computer_Vision.md

@@ -0,0 +1,8 @@
+### Papers
+
+- Snorkel: Rapid Training Data Creation with Weak Supervision (2017. 11)
+	- **`Labelling Functions`**, **`Data Programming`**
+	- [arXiv](https://arxiv.org/abs/1711.10160), [the_morning_blog](https://blog.acolyer.org/2018/08/22/snorkel-rapid-training-data-creation-with-weak-supervision/)
+- Training classifiers with natural language explanations (2018. 5)
+	- **`Babble Labble`**, **`Data Programming`**
+	- [arXiv](https://arxiv.org/abs/1805.03818), [the_morning_blog](https://blog.acolyer.org/2018/08/24/training-classifiers-with-natural-language-explanations/)

papers/Framework_System.md

@@ -0,0 +1,104 @@
+### Papers
+
+- **Dropout** (2012, 2014)
+	- **`Regulaizer`**, **`Ensemble`**
+	- [arXiv (2012)](https://arxiv.org/abs/1207.0580), [arXiv (2014)](https://www.cs.toronto.edu/~hinton/absps/JMLRdropout.pdf), [note](notes/dropout.md)
+- Regularization of Neural Networks using DropConnect (2013)
+	- **`Regulaizer`**, **`Ensemble`**
+	- [paper](https://cs.nyu.edu/~wanli/dropc/dropc.pdf), [note](notes/dropconnect.md), [wanli_summary](https://cs.nyu.edu/~wanli/dropc/)
+- Recurrent Neural Network Regularization (2014. 9)
+	- **`RNN`**, **`Dropout to Non-Recurrent Connections`**
+	- [arXiv](https://arxiv.org/abs/1409.2329)
+- **Batch Normalization** (2015. 2)
+	- **`Regulaizer`**, **`Accelerate Training`**, **`CNN`**
+	- [arXiv](https://arxiv.org/abs/1502.03167), [note](notes/batch_normalization.md)
+- Training Very Deep Networks (2015. 7)
+	- **`Highway`**, **`LSTM-like`**
+	- [arXiv](https://arxiv.org/abs/1507.06228), [note](notes/highway_networks.md)
+- A Theoretically Grounded Application of Dropout in Recurrent Neural Networks (2015. 12)
+	- **`Variational RNN`**, **`Dropout - RNN`**, **`Bayesian interpretation`**
+	- [arXiv](https://arxiv.org/abs/1512.05287)
+- Deep Networks with Stochastic Depth (2016. 3)
+	- **`Dropout`**, **`Ensenble`**, **`Beyond 1000 layers`**
+	- [arXiv](https://arxiv.org/abs/1603.09382), [note](notes/stochastic_depth.md)
+- Adaptive Computation Time for Recurrent Neural Networks (2016. 3)
+	- **`ACT`**, **`Dynamically`**, **`Logic Task`**
+	- [arXiv](https://arxiv.org/abs/1603.08983)
+- Layer Normalization (2016. 7)
+	- **`Regulaizer`**, **`Accelerate Training`**, **`RNN`**
+	- [arXiv](https://arxiv.org/abs/1607.06450), [note](notes/layer_normalization.md)
+- Recurrent Highway Networks (2016. 7)
+	- **`RHN`**, **`Highway`**, **`Depth`**, **`RNN`**
+	- [arXiv](https://arxiv.org/abs/1607.03474), [note](notes/recurrent_highway.md)
+- Using Fast Weights to Attend to the Recent Past (2016. 10)
+	- **`Cognitive`**, **`Attention`**, **`Memory`**
+	- [arXiv](https://arxiv.org/abs/1610.06258), [note](notes/fast_weights_attn.md)
+- Professor Forcing: A New Algorithm for Training Recurrent Networks (2016. 10)
+	- **`Professor Forcing`**, **`RNN`**, **`Inference Problem`**, **`Training with GAN`**
+	- [arXiv](https://arxiv.org/abs/1610.09038), [note](notes/professor_forcing.md)
+- Equality of Opportunity in Supervised Learning (2016. 10)
+	- **`Equalized Odds`**, **`Demographic Parity`**, **`Bias`**
+	- [arXiv](https://arxiv.org/abs/1610.02413), [the_morning_paper](https://blog.acolyer.org/2018/05/07/equality-of-opportunity-in-supervised-learning/)
+- Categorical Reparameterization with Gumbel-Softmax (2016. 11)
+	- **`Gumbel-Softmax distribution `**, **`Reparameterization`**, **`Smooth relaxation`**
+	- [arXiv](https://arxiv.org/abs/1611.01144), [open_review](https://openreview.net/forum?id=rkE3y85ee)
+- Understanding deep learning requires rethinking generalization (2016. 11)
+	- **`Generalization Error`**, **`Role of Regularization`**
+	- [arXiv](https://arxiv.org/abs/1611.03530)
+- Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer (2017. 1)
+	- **`MoE Layer`**, **`Sparsely-Gated`**, **`Capacity`**, **`Google Brain`**
+	- [arXiv](https://arxiv.org/abs/1701.06538), [note](notes/very_large_nn_moe_layer.md)
+- **A simple neural network module for relational reasoning** (2017. 6)
+	- **`Relational Reasoning`**, **`DeepMind`**
+	- [arXiv](https://arxiv.org/abs/1706.01427), [note](notes/relational_network.md), [code](https://github.com/DongjunLee/relation-network-tensorflow)
+- On Calibration of Modern Neural Networks (2017. 6)
+	- **`Confidence calibration`**, **`Maximum Calibration Error (MCE)`**
+	- [arXiv](https://arxiv.org/abs/1706.04599)
+- When is a Convolutional Filter Easy To Learn? (2017. 9)
+	- **`Conv + ReLU`**, **`Non-Gaussian Case`**, **`Polynomial Time`**
+	- [arXiv](https://arxiv.org/abs/1709.06129), [open_review](https://openreview.net/forum?id=SkA-IE06W)
+- mixup: Beyond Empirical Risk Minimization (2017. 10)
+	- **`Data Augmentation`**, **`Vicinal Risk Minimization`**, **`Generalization`**
+	- [arXiv](https://arxiv.org/abs/1710.09412), [open_review](https://openreview.net/forum?id=r1Ddp1-Rb)
+- Measuring the tendency of CNNs to Learn Surface Statistical Regularities (2017. 11)
+	- **`not learn High Level Semantics`**, **`learn Surface Statistical Regularities`**
+	- [arXiv](https://arxiv.org/abs/1711.11561), [the_morning_paper](https://blog.acolyer.org/2018/05/29/measuring-the-tendency-of-cnns-to-learn-surface-statistical-regularities/)
+- MentorNet: Regularizing Very Deep Neural Networks on Corrupted Labels (2017. 12)
+	- **`MentorNet - StudentNet`**, **`Curriculum Learning`**, **`Output is Weight`**
+	- [arXiv](https://arxiv.org/abs/1712.05055)
+- Deep Learning Scaling is Predictable, Empirically (2017. 12)
+	- **`Power-Law Exponents`**, **`Grow Training Sets`**
+	- [arXiv](https://arxiv.org/abs/1712.00409), [the_morning_paper](https://blog.acolyer.org/2018/03/28/deep-learning-scaling-is-predictable-empirically/)
+- Sensitivity and Generalization in Neural Networks: an Empirical Study (2018. 2)
+	- **`Robustness`**, **`Data Perturbations`**, **`Survey`**
+	- [arXiv](https://arxiv.org/abs/1802.08760), [open_review](https://openreview.net/forum?id=HJC2SzZCW)
+- Can recurrent neural networks warp time? (2018. 2)
+	- **`RNN`**, **`Learnable Gate`**, **`Chrono Initialization`**
+	- [open_review](https://openreview.net/forum?id=SJcKhk-Ab)
+- Spectral Normalization for Generative Adversarial Networks (2018. 2)
+	- **`GAN`**, **`Training Discriminator`**, **`Constrain Lipschitz`**, **`Power Method`**
+	- [open_review](https://openreview.net/forum?id=B1QRgziT-&noteId=BkxnM1TrM)
+- On the importance of single directions for generalization (2018. 3)
+	- **`Importance`**, **`Confusiing Neurons`**, **`Selective Neuron`**, **`DeepMind`**
+	- [arXiv](https://arxiv.org/abs/1803.06959), [deepmind_blog](https://deepmind.com/blog/understanding-deep-learning-through-neuron-deletion/)
+- Group Normalization (2018. 3)
+	- **`Group Normalization (GN)`**, **`Batch (BN)`**, **`Layer (LN)`**, **`Instance (IN)`**, **`Independent Batch Size`**
+	- [arXiv](https://arxiv.org/abs/1803.08494)
+- Fast Decoding in Sequence Models using Discrete Latent Variables (2018. 3)
+	- **`Autoregressive`**, **`Latent Transformer`**, **`Discretization`**
+	- [arXiv](https://arxiv.org/abs/1803.03382)
+- Delayed Impact of Fair Machine Learning (2018. 3)
+	- **`Outcome Curve`**, **`Max Profit, Demographic Parity, Equal Opportunity`**
+	- [arXiv](https://arxiv.org/abs/1803.04383), [the_morning_paper](https://blog.acolyer.org/2018/08/13/delayed-impact-of-fair-machine-learning/), [bair_blog](https://bair.berkeley.edu/blog/2018/05/17/delayed-impact/)
+- How Does Batch Normalization Help Optimization? (No, It Is Not About Internal Covariate Shift) (2018. 5)
+	- **`Smoothing Effect`**, **`BatchNorm’s Reparametrization`**
+	- [arXiv](https://arxiv.org/abs/1805.11604)
+- When Recurrent Models Don't Need To Be Recurrent (2018. 5)
+	- **`Approximate`**, **`Feed-Forward`**
+	- [arXiv](https://arxiv.org/abs/1805.10369), [bair_blog](http://bair.berkeley.edu/blog/2018/08/06/recurrent/)
+- Relational inductive biases, deep learning, and graph networks (2018, 6)
+	- **`Survey`**, **`Relation`**, **`Graph`**
+	- [arXiv](https://arxiv.org/abs/1806.01261)
+- Universal Transformers (2018. 7)
+	- **`Transformer`**, **`Weight Sharing`**, **`Adaptive Computation Time (ACT)`**
+	- [arXiv](https://arxiv.org/abs/1807.03819), [google_ai_blog](https://ai.googleblog.com/2018/08/moving-beyond-translation-with.html)