Thursday 9:10-12:10 (BL Building 113) [pdf]
Event | Date | Description | Course Material |
---|---|---|---|
Lecture | 09/22/2016 | Introduction | Slides: [Course Logistics] [Introduction] |
Lecture | 09/29/2016 | Neural Network | Suggested Readings: 1. [Linear Algebra] [Linear Algebra Slides] 2. [Linear Algebra Quick Review] Slides: [NN Basics] |
Lecture | 10/06/2016 | Backpropagation Word Representation |
Suggested Readings: 1. [Learning Representations by Backpropogating Errors] 2. [From Frequency to Meaning: Vector Space Models of Semantics] Slides: [Backpropagation] [Word Representation 1] |
Lecture | 10/13/2016 | Word Representation |
Suggested Readings: 1. [Distributed Representations of Words and Phrases and their Compositionality] 2. [Efficient Estimation of Word Representations in Vector Space] 3. [GloVe: Global Vectors for Word Representation] Slides: [Word Representation 2] [TensorFlow Tutorial] |
A1 Release | 10/13/2016 | Word Embedding | [A1 Page] [A1 Slides] |
Lecture | 10/20/2016 | Sequence Modeling Recursive Neural Network |
Suggested Readings: 1. [Parsing with Compositional Vector Grammars] 2. [Subgradient Methods for Structure Prediction] 3. [Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank] Slides: [Sequence Modeling] [Recursive NN] |
Lecture | 10/27/2016 | Recurrent Neural Network |
Suggested Readings: 1. [RNNLM: Recurrent Neural Nnetwork Language Model] 2. [Extensions of RNNLM] 3. [RNN for Language Understanding] 4. [RNN for Joint Language Understanding] 5. [Sequence-to-Sequence Learning] 6. [Neural Conversational Model] Slides: [Recurrent NN 1] |
A2 Release | 10/27/2016 | Sentiment Analysis | [A2 Page] [A2 Slides] |
A1 Due | 10/30/2016 | Word Embedding | |
Guest Lecture | 11/03/2016 | Guest Lecture by HTC Convolutional Neural Network |
Slides: [Convolutional NN] |
Lecture | 11/10/2016 | NN Practical Tips Recurrent Neural Network |
Suggested Readings: 1. [Initialize Recurrent Networks of Rectified Linear Units] 2. [Practical recommendations for gradient-based training of deep architectures] Slides: [Company Event] [NN Practical Tips] [Recurrent NN 2] |
A3 Release | 11/10/2016 | Natural Language Understanding |
[A3 Page]
[A3 Slides] Suggested Readings: 1. [RNN for Language Understanding] 2. [RNN for Joint Language Understanding] |
A2 Due | 11/17/2016 | Sentiment Analysis | |
Special | 11/17/2016 | Company Workshop by Yahoo! [Yahoo! Data Engineering Project Intern] [Yahoo! Data Scientist Full Time Employee] |
[Confirmed Registration List]
09:00 - 09:30 Event Checkin 09:30 - 09:40 Welcome Intro 09:40 - 10:20 Data Engineering Team Sharing 10:30 - 11:10 Product Engineering Team Sharing 11:10 - 11:40 Search Team Sharing 11:40 - 12:00 Group Photo and Good Bye |
Lecture | 11/24/2016 | Gating Mechanism Attention and Memory |
Suggested Readings: 1. [Long Short-Term Memory] 2. [Gated Recurrent Unit] 3. [Neural Machine Translation with Attention] 4. [Summarization with Attention] Slides: [Gating Mechanism] [Attention and Memory] |
Project Announcement | 11/24/2016 | P1: Dialogue State Tracking Challenge P2: Machine Comprehension |
[P1 Page]
[P1 Slides] [P2 Page] [P2 Slides] |
Lecture | 12/01/2016 | Deep Reinforcement Learning |
Suggested Readings: 1. [Stephane Ross’ thesis (Introduction)] Slides: [Deep Reinforcement Learning 1] |
A3 Due | 12/01/2016 | Natural Language Understanding | |
A4 Release | 12/01/2016 | Language Generation |
[A4 Page]
[A4 Slides] Suggested Readings: 1. [Sequence to Sequence Machine Translation] 2. [Semantically Conditioned Natural Language Generation] [Semantically Conditioned NLG Slides] |
Lecture | 12/08/2016 | Deep Reinforcement Learning |
Suggested Readings: 1. [Playing Atari with Deep Reinforcement Learning] 2. [Continuous Control with Deep Reinforcement Learning] Slides: [Deep Reinforcement Learning 2] |
A5 Release | 12/08/2016 | Game Playing |
[A5 Page]
[A5 Slides] |
A4 Due | 12/15/2016 | Language Generation | |
Special | 12/15/2016 | Company Workshop by Microsoft | [Confirmed Registration List]
09:00 - 09:30 Opening 09:30 - 10:00 Internship and Employment Program Sharing 10:00 - 10:30 Future Vision of Technology 10:30 - 11:30 Tech Talk (Machine Learning) 11:30 - 12:00 Company Tour |
Lecture | 12/22/2016 | Unsupervised Learning |
Suggested Readings: 1. [Generative Adversarial Nets] 2. [Deep Convolutional Generative Adversarial Nets] Slides: [Dialogue System] [Machine Comprehension] [Unsupervised Learning] |
Lecture | 12/29/2016 | Generative Model/Deep Learning Trends |
Suggested Readings: 1. [Conditional GAN] 2. [Adversarially Learned Inference] 3. [Improved Techniques for Training GANs] 4. [Unrolled GANs] 5. [Generative Adversarial Parallelization] Slides: [Generative Model] [Deep Learning Trends] [Review] |
A5 Due | 12/29/2016 | Game Playing | |
Special | 01/05/2016 | Break | |
Project Due | 01/08/2017 | Dialogue State Tracking Challenge and Machine Comprehension | |
Project Presentation | 01/12/2017 | Final Project Oral Presentation |