Courses

This course introduces core deep learning techniques used in modern natural language processing and visits recent advances and the challenges unresolved.
 

This course covers the following topics:

• How NLP utilizes representation learning.

How problems in NLP motivated novel deep learning approaches such as attention and self-attention.

Architectures: RNN, Sequence-to-sequence model, Transformer.

Applications: Word Embedding, Language Model, Core NLP (Parsing, SRL), QA and Natural Language Generation (Summarization, Machine Translation, etc.) problems.

Recent Concerns: Controllability, Injection of Constraints and Knowledge, Bias & Fairness

This course introduces up-to-date research trends in various data science-related fields. Students will learn specialized topics and new methodologies through lectures and seminars.

This course covers the following topics:

• Hallucination

• Knowledge Conflict

• Fact Verification

• Retrieval Augmented Generation

• Prompting

• Open-domain Question Answering

All areas of data science are concerned with collecting and analyzing data. This course is designed to provide foundations of probability and statistics. From this course, student will understand how probability and statistics explain the data generating process and can be used to analyze data. 
 

This course covers the following topics:

• Probability 

• Random Variables 

• Expectation 

• Convergence of Random Variables 

• Statistical Inference 

• Non-parametric method (such as Bootstrap) 

• Parametric Inference 

• Hypothesis test 

• Bayesian Inference