machine learning seminar: Yuan -- Optimizing LLMs and Optimizing with LLMs

The next ML seminar will be given by Prof Xiaoming Yuan from the department of
mathematics, university of Hong Kong.  the details are given below: 

When and where: 1pm, September 15 (Monday) in AGR room (Carslaw building 829) 

Title: Optimizing LLMs and Optimizing with LLMs 



Abstract: Large Language Models (LLMs) have brought about a profound revolution across
various sectors.  While the advancement of LLMs has primarily relied on hardware such as
GPUs and engineering technologies, the field is now entering a new phase that
necessitates deeper engagement with scientific disciplines.  The integration of
mathematical and quantitative insights is crucial for driving the next wave of
breakthroughs in the realm of LLMs.  We will reevaluate the core tasks involved in the
lifecycle of LLMs through an optimization lens, with a focus on enhancing the scientific
understanding of key phases including pre-training and post-training.  We will discuss
specific tasks such as (distributed) low-bit training, pruning, and quantization.  Our
primary objective is to reduce the computation and memory footprint of LLMs throughout
their lifecycle.  We will also initiate two new concepts: Optimization Agents and
Intelligence-Collaborative Computation, for the next generation of computation.