Artificial Intelligence

• A Roadmap towards Machine Intelligence
• Building Machines That Learn and Think Like People
• What Question Would Turing Pose Today?

Techniques

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• The Difficulty of Training Deep Architectures and the Effect of Unsupervised Pre-Training
• Advances In Optimizing Recurrent Networks
• Recurrent Neural Network Regularization
• Dynamic Capacity Networks
• Net2Net: Accelerating Learning via Knowledge Transfer
• Recurrent Batch Normalization

Computer Vision

Object Classification and Recognition

• ImageNet Classification with Deep Convolutional Neural Networks
• Going Deeper with Convolutions
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Visualizing and Understanding Convolutional Networks
• Intriguing Properties of Neural Networks
• Deep Residual Learning for Image Recognition
• Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps

Captioning, Style-Transfer and other extensions

• Perceptual Losses for Real-Time Style Transfer and Super-Resolution
• DenseCap: Fully Convolutional Localization Networks for Dense Captioning
• Network in Network
• Residual Networks are Exponential Ensembles of Relatively Shallow Networks
• Residual Networks of Residual Networks: Multilevel Residual Networks
• Identity Mappings in Deep Residual Networks
• Learning Deep Features for Scene Recognition using Places Database

Image Generation

• DRAW: A Recurrent Neural Network For Image Generation
• Generative Adversarial Nets
• Conditional Generative Adversarial Nets
• Deep Convolutional Generative Adversarial Nets
• Conditional Image Generation with PixelCNN Decoders
• Pixel Recurrent Neural Network

Videos

• Delving Deeper into Convolutional Networks for Learning Video Representations
• Actions ~ Transformations

Visual Question Answering

• Neural Module Networks
• Stacked Attention Networks for Image Question Answering

Other

• You Only Look Once: Unified, Real-Time Object Detection
• Deep Visual Analogy-Making
• Object Detectors Emerge in Deep Scene CNNs
• Deep Networks with Stochastic Depth
• How transferable are features in deep neural networks?

Natural Language Processing

Embeddings

• GloVe: Global Vectors for Word Representation
• Improving Word Representations via Global Context and Multiple Word Prototypes
• Efficient Estimation of Word Representations in Vector Space
• Skip-Thought Vectors

Reading and comprehension

• WikiReading : A Novel Large-scale Language Understanding Task over Wikipedia
• Teaching Machines to Read and Comprehend

Translation

• Sequence to Sequence Learning with Neural Networks
• Neural Machine Translation by Jointly Learning to Align and Translate
• Fully Character-Level Neural Machine Translation without Explicit Segmentation
• Addressing the Rare Word Problem in Neural Machine Translation
• Achieving Open Vocabulary Neural Machine Translation with Hybrid Word-Character Models
• Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation

Question Answering

• WikiQA: A challenge dataset for open-domain question answering
• Towards AI-Complete Question Answering: A Set of Prerequisite Toy Tasks
• Question Answering with Subgraph Embeddings

Conversation / Dialog

• A Neural Conversational Model
• A Persona-Based Neural Conversation Model
• Evaluating Prerequisite Qualities for Learning End-to-end Dialog Systems
• Smart Reply: Automated Response Suggestion for Email
• Deep Reinforcement Learning for Dialogue Generation

Other

• Bag of Tricks for Efficient Text Classification (critique)
• Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer
• Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank
• Convolutional Neural Network For Sentence Classification

Reinforcement learning

• Human-Level Control Through Deep Reinforcement Learning
• Bridging the Gap Between Value and Policy Based Reinforcement Learning
• Deep Reinforcement Learning for Dialogue Generation
• [Technical Talk Notes] Deep Reinforcement Learning via Policy Optimization by John Schulman of OpenAI
• Dynamic Frame skip Deep Q Network

Code generation, or learning a computer program

• Learning To Execute
• Neural Turing Machines
• Neural Generation of Regular Expressions from Natural Language with Minimal Domain Knowledge

Visualization

• Visualizing Data using t-SNE
• Visualizing and Understanding Convolutional Networks

Other

• Deep Sequence Models for Premise Selection

Last updated: 8th March, 2017