Titile: How to learn language models and how language models can learn
Date: 2024-06-28 11:00-12:00
Location: CSIE R102
Speaker: Prof. Cho-Jui Hsieh,UCLA
Host: Prof. Chih-Jen Lin
Abstract:
How do we learn from data? In the first part of the talk, we will discuss the traditional way, where optimizers are applied to learn the model weights from data. We will delve into a range of higher-order optimization algorithms, as well as a novel framework to design optimizers automatically. The second part of the presentation will introduce a new learning paradigm, showcasing how Large Language Models (LLMs) can learn fully interpretable models from data, and the learning in this case is done by the LLM itself.
Biography:
Cho-Jui Hsieh is an associate professor in the Computer Science Department at UCLA and a research scientist at Google. His work primarily focuses on enhancing the efficiency and robustness of machine learning systems, and he has made significant contributions to multiple widely-used machine learning packages. He has been honored with the NSF Career Award, Samsung AI Researcher of the Year, and Google Research Scholar Award, and his work has been acknowledged with several paper awards in ICLR, KDD, ICDM, ICPP, and SC.