CS291K - Schedule home | schedule


 

Date

Topic

Comments

Apr 2 Course Introduction  
Apr 4 Neural Network Basic Concepts
Apr 9 Guest Speaker: Andrew Mutz for course project topics  
Apr 11 Neural Network Training + CNN  
Apr 13 RNN + LSTM Friday make-up
Apr 16 Word Embedding
Distributed Representations of Words and Phrases and their Compositionality - NIPS'13
Deep contextualized word representations - NAACL'18

Optional:
GloVe: Global Vectors for Word Representation - EMNLP'14
A Neural Probabilistic Language Model - JMLR'03
Indexing by Latent Semantic Analysis - JASIS'90
Basic Concepts + Advanced Topics
Apr 18 GAN
Generative Adversarial Nets, NIPS'14
MaskGAN: Better Text Generation via Filling in the ___, ICLR'18

Optional:
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks, ICLR'16
SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient, AAAI'17
Adversarial Generation of Natural Language, ACL Representation Learning Workshop'17s
 
Apr 20 Variational Auto-Encoder
Auto-Encoding Variational Bayes
Variational Autoencoders Explained

Optional:
Variational Inference: A Review for Statisticians
Semi-Amortized Variational Autoencoders
Friday make-up
Apr 23 No Class  
Apr 25 No Class  
Apr 30 Reinforcement Learning
Deep Reinforcement Learning for Dialogue Generation (EMNLP'16)
A Deep Reinforced Model for Abstractive Summarization (ICLR'18)

Optional:
Sequence Level Training with Recurrent Neural Networks (ICLR'16)
DCN+: MIXED OBJECTIVE AND DEEP RESIDUAL COATTENTION FOR QUESTION ANSWERING (ICLR'18)
Ranking Sentences for Extractive Summarization with Reinforcement Learning (Arxiv'18)
Project Proposal Due
Advanced Topics
May 2 Attention
Convolutional and multi-hop attention
Attention is all you need

Optional:
QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehension
The first paper review submission
May 7 No Class  
May 9 Deep Networks
ImageNet Classification with Deep Convolutional Neural Networks
Deep Residual Learning for Image Recognition
Neural Architectures for Named Entity Recognition

Optional:
Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning
May 11 Sequence2Sequence
Sequence to sequence learning with neural networks
Pointer networks
CTC speech recognition sequence modeling

Friday make-up
May 14 Parsing
A Fast and Accurate Dependency Parser using Neural Networks
Transition-Based Dependency Parsing with Stack Long Short-Term Memory
DEEP BIAFFINE ATTENTION FOR NEURAL DEPENDENCY PARSING

 
May 16 Dialog
How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generation (EMNLP'16)
Towards an Automatic Turing Test: Learning to Evaluate Dialogue Responses
A Neural Network Approach to Context-Sensitive Generation of Conversational Responses
Adversarial Learning for Neural Dialogue Generation

 
May 21 Meta Learning
Model-agnostic meta-learning for fast adaptation of deep networks
Meta-learning with memory-augmented neural networks
Learning to learn by gradient descent by gradient descent

Optional:
Building machines that learn and think like people
 
May 23 Capsule network
Dynamic Routing Between Capsules
Capsule Network Performance on Complex Data

Optional:
Matrix Capsules with EM routing
Transforming auto-encoders
Untangling invariant object recognition

Zero-Shot Learning
Zero-Shot Learning - A Comprehensive Evaluation of the Good, the Bad and the Ugly
The second paper review submission
May 28 Memorial Day  
May 30 Transfer Learning
How transferable are features in deep neural networks?

Optional:
How Transferable are Neural Networks in NLP Applications?
June 4 Project Presentation  
June 6 Project Presentation  
June 8 Project Presentation Overflow June 11 as the backup
June 13 Project Final Report Due