Self-Imitation Learning of Locomotion Movements through Termination Curriculum

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Researchers

Research units

Abstract

Animation and machine learning research have shown great advancements in the past decade, leading to robust and powerful methods for learning complex physically-based animations. However, learning can take hours or days, especially if no reference movement data is available. In this paper, we propose and evaluate a novel combination of techniques for accelerating the learning of stable locomotion movements through self-imitation learning of synthetic animations. First, we produce synthetic and cyclic reference movement using a recent online tree search approach that can discover stable walking gaits in a few minutes. This allows us to use reinforcement learning with Reference State Initialization (RSI) to find a neural network controller for imitating the synthesized reference motion. We further accelerate the learning using a novel curriculum learning approach called Termination Curriculum (TC), that adapts the episode termination threshold over time. The combination of the RSI and TC ensures that simulation budget is not wasted in regions of the state space not visited by the final policy. As a result, our agents can learn locomotion skills in just a few hours on a modest 4-core computer. We demonstrate this by producing locomotion movements for a variety of characters.

Details

Original languageEnglish
Title of host publicationProceedings - MIG 2019
Subtitle of host publicationACM Conference on Motion, Interaction, and Games
Publication statusPublished - 31 Oct 2019
MoE publication typeA4 Article in a conference publication
EventACM SIGGRAPH Conference on Motion, Interaction and Games - Newcastle Upon Tyne, United Kingdom
Duration: 28 Oct 201930 Oct 2019
Conference number: 12

Conference

ConferenceACM SIGGRAPH Conference on Motion, Interaction and Games
Abbreviated titleMIG
CountryUnited Kingdom
CityNewcastle Upon Tyne
Period28/10/201930/10/2019

ID: 38449274