XAI630: Meta-Learning

Information

Outline

Recent developments in artificial intelligence research are moving towards Artificial General Intelligence (AGI), where a single AI agent can simultaneously perform multiple tasks while learning and evolving on its own. In line with these current AI research trends, this course will introduce an overview of meta-learning which is a key research area for realizing AGI. In specific, this course will cover important meta-learning algorithms that allow an AI model or agent to automatically and quickly adapt to new tasks.

Objective

Throughout the course, students will learn the backgrounds and various algorithms of meta learning and understand key factors in meta learning. In particular, the course will cover basic meta learning problems and algorithms, especially in comparison to multi-task learning, transfer learning, few-shot learning, continual learning, and automated machine learning. On top of that, this course will introduce some advanced topics such as meta reinforcement learning, neural architecture search, auto-augment, in-context learning, and representation learning. The course aims students to have enough knowledge and moreover an insight for the corresponding researches.

Pre-requisites

Basic knowledges in machine learning and deep learning are required.

Material

Lecture notes will be the main material of the course, and these do not come from a single textbook. However, the following reference text books will be useful.

Grading

Schedule