About The Webinar

The classic machine learning paradigm learns in isolation, which is only suitable for well-defined narrow tasks in closed environments. It is far from sufficient for truly intelligent systems, which must learn continually, accumulate the knowledge learned in the past, and selectively transfer the knowledge to help learn each new task. The machine learning paradigm, continual/lifelong learning, aims to achieve these goals. In this talk, I will first introduce the context of and motivation for continual learning and then discuss some of our recent work on continual learning of natural language processing tasks.

Past webinars

Some of our speakers

Speaker 1

RADA MIHALCEA

University of Michigan

Speaker 1

ROBERTO NAVIGLI

University of Rome

Speaker 1

CHRISTIANE FELLBAUM

Princeton University

Speaker 1

BING LIU

University of Illinois