Deep Reinforcement Learning with Human Intervention Applications

Speaker:  Zehong (Jimmy) Cao – Adelaide, SA, Australia
Topic(s):  Human Computer Interaction

Abstract

Deep reinforcement learning (DRL) considers the problem of machine agent learning to make decisions by trial and error and make decisions to maximise rewards by continuously interacting with the environment. DRL has been used for a diverse set of applications, such as robotics, games, transportation, and healthcare. In this talk, the fundamentals of DRL will be introduced and how to design a single agent or multi-agent system using Unity platform to fit the user needs. Then, to interact with human knowledge, the reward learning from human preferences and behaviours will be presented used to resolve complex DRL tasks, such as MuJoCo games and physical robot controls. The naturally inspired human perceptions are beneficial for precise reward design and can be applied to state-of-the-art RL systems with human-in-the-loop learning, such as human-autonomy teaming systems. In addition, the DRL for autonomous driving in transportation applications will be demonstrated and explores a solution for driving safety

About this Lecture

Number of Slides:  30
Duration:  60 minutes
Languages Available:  Chinese (Simplified), English
Last Updated: 

Request this Lecture

To request this particular lecture, please complete this online form.

Request a Tour

To request a tour with this speaker, please complete this online form.

All requests will be sent to ACM headquarters for review.