CIS 607: Seminar on Deep Learning for Natural Language Processing


Course Description

Deep learning has recently proved itself as a powerful branch of machine learning that has applications in a variety of domains (i.e., computer vision, robotics, etc.). In natural language processing (NLP), deep learning has been transformative and led to a new generation of methods with better performance, portability and robustness (e.g, the machine translation systems). All of those advances are very recent and the demand for data scientists with deep learning expertise is growing very quickly. At the beginning of this seminar, we will cover the basic concepts of deep learning and NLP, and possibly provide some hand-on experience to implement the models. Afterward, we will review and discuss a collection of research papers on NLP with deep learning, including but not limited to the typical tasks of question answering, information extraction, natural language inference, dialog and summarization.

Prerequisites: Some background in algebra and machine learning (i.e., CIS 472/572). Please talk with the instructor if you are not sure if you have the necessary background.

Class Location: 200 Deschutes Hall

Class Time: Friday, 14:30 pm to 15:50 pm

Instructor: Thien Huu Nguyen
Email: thien AT cs.uoregon.edu
Office: Room 330, Deschutes Hall

Course Material:
Goldberg. Neural Network Methods for Natural Language Processing 2017. (free download when on UO campus).
Kyunghyun Cho's note
ACL Anthology (free and covering the papers in most of the major NLP venues)

Class Organization
The grading for this class is P/NP.
Each student in the class will need to choose some paper on some particular topic, present them in one of the classes and lead the discussion on the topic.
Each presentation will be given 30 minutes along with 5-10 minutes for discussions. More discussion is encouraged on Piazza.
After the presentation, the presenter needs to submit a summary about the presented paper/topic. The summary should follow the NAACL format (a.k.a. ACL style for 8.5x11 paper). The required length of the summary is between 2 and 3 pages.
For each presentation, we will have one student (other than the presenter) to serve as the reviewer. The role of the reviewer is to provide judgement/comments/suggestion or ask questions about the paper/topic in the discussion time after the presentation. Although all students need to read and understand the papers being presented before each class to be able to actively contribute to the discussions, the reviewer would help to provide deeper judgement by reading and thinking critically about the papers/topics ahead of time.

IMPORTANT:

Please select a paper you want to present in this list. Write your name next to the paper you select. You are welcome to choose another paper that is not in the list. Please talk with the instructor if you want to do this. Also, please provide the ID of the paper/topic you want to review next to your name. All the paper assignments should be done before Jan 15 (no later than that) so we can schedule the presentations.

Tentative Schedule

Please sign up on Piazza for discussions.
Week Topics Presenter Slides Reviewer
01/10 Introduction to Modern Natural Language Processing Thien Link
01/17 Deep RNNs Encode Soft Hierarchical Syntax (ACL 2018) Amir Pouran Ben Veyseh Link Mohammad Eshghi
Amir's Recent Work on Relation Extraction Amir Pouran Ben Veyseh Link
01/24 Under the Hood: Using Diagnostic Classifiers to Investigate and Improve how Language Models Track Agreement Information (EMNLP 2018) Ziyad Alsaeed Link Adam Noack
Tracking State Changes in Procedural Text: A Challenge Dataset and Models for Process Paragraph Comprehension (NAACL 2018) Avery Scheiwiller Link Andrew Butler
01/31 What you can cram into a single $&!#* vector: Probing sentence embeddings for linguistic properties (ACL 2018) Shravan Kale Link Qiuhao Lu
Multimodal Compact Bilinear Poolingfor Visual Question Answering and Visual Grounding (EMNLP 2016) Mohammad Eshghi Link Avery Scheiwiller
02/07 No Class (AAAI 2020)
02/14 Dynamic Integration of Background Knowledge in Neural NLU Systems Qiuhao Lu Link Amir Pouran Ben Veyseh
Training Classifiers with Natural Language Explanations (ACL 2018) Adam Noack Link Shravan Kale
02/21 How Much Attention Do You Need?A Granular Analysis of Neural Machine Translation Architectures (ACL 2018) Steven Walton Link Sudharshan Srinivasan
Visualizing and Understanding the Effectiveness of BERT (EMNLP 2019) Andrew Butler Link Haoran Wang
02/28 Fine-Grained Temporal Relation Extraction (ACL 2019) Haoran Wang Link Abhishek Yenpure
Joint Type Inference on Entities and Relations via Graph Convolutional Networks (ACL 2019) Zayd Hammoudeh Link Steven Walton
03/06 A Mutual Information Maximization Perspective of Language Representation Learning (ICLR 2020) Abhishek Yenpure Link Zayd Hammoudeh
Graph Transformer Networks (NIPS 2019) Sudharshan Srinivasan Link Ziyad Alsaeed
03/13 Universal Language Model Fine-tuning for Text Classification (ACL 2018) Parsa Bagheri Link -
Research discussion and additional presentations