Scaling up Deep Reinforcement Learning for Intelligent Video Game Agents

Anton Debner*

*Corresponding author for this work

Research output: Contribution to conferenceAbstractScientificpeer-review

3 Citations (Scopus)

Abstract

We introduce the concept of Deep Reinforcement Learning (DRL) and describe the current state-of-the-art in subareas relevant to the author's research. We present previous and ongoing work done by the author in context of game engines, video game development and Machine Learning performance. We discuss our measurements showing the performance discrepancy between training DRL agents on game engines and end-to-end GPU-based physics simulators. We propose the use of external GPU-based physics simulators and transfer learning to accelerate the training of DRL models for game engines. As future work, we discuss the use of model decomposition in complex environments to further accelerate learning efficiency of DRL in addition to increased hardware utilization.

Original languageEnglish
Pages192-193
Number of pages2
DOIs
Publication statusPublished - 2022
MoE publication typeNot Eligible
EventIEEE International Conference on Smart Computing - Espoo, Finland
Duration: 20 Jun 202224 Jun 2022
Conference number: 8

Conference

ConferenceIEEE International Conference on Smart Computing
Abbreviated titleSMARTCOMP
Country/TerritoryFinland
CityEspoo
Period20/06/202224/06/2022

Keywords

  • deep learning
  • deep reinforcement learning
  • game development
  • game engines
  • performance

Fingerprint

Dive into the research topics of 'Scaling up Deep Reinforcement Learning for Intelligent Video Game Agents'. Together they form a unique fingerprint.

Cite this