Beyond Discrete Environments: Benchmarking Regret-Based Automatic Curriculum Learning in MuJoCo

Chin-Jui Chang, Chen-Xing Li, Chen-Xing Li, Jan Seyler, Shahram Eivazi, Shahram Eivazi

2025

Abstract

Training robust reinforcement learning (RL) agents capable of performing well in unseen scenarios remains a significant challenge. Curriculum learning has emerged as a promising approach to build transferable skills and enhance overall robustness. This paper investigates regret-based adversarial methods for automatically generating curricula, extending their evaluation beyond simple environments to the more complex MuJoCo suite. We benchmark several state-of-the-art regret-based methods against traditional baselines, revealing that while these methods generally outperform baselines, the performance gains are less substantial than anticipated in these more complex environments. Moreover, our study provides valuable insights into the application of regret-based curriculum learning methods to continuous parameter spaces and highlights the challenges involved. We discuss promising directions for improvement and offer perspectives on how current automatic curriculum learning techniques can be applied to real-world tasks.

Paper Citation

in Harvard Style

Chang C., Li C., Seyler J. and Eivazi S. (2025). Beyond Discrete Environments: Benchmarking Regret-Based Automatic Curriculum Learning in MuJoCo. In Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART; ISBN 978-989-758-737-5, SciTePress, pages 133-143. DOI: 10.5220/0013116400003890

in Bibtex Style

@conference{icaart25,
author={Chin-Jui Chang and Chen-Xing Li and Jan Seyler and Shahram Eivazi},
title={Beyond Discrete Environments: Benchmarking Regret-Based Automatic Curriculum Learning in MuJoCo},
booktitle={Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART},
year={2025},
pages={133-143},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013116400003890},
isbn={978-989-758-737-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART
TI - Beyond Discrete Environments: Benchmarking Regret-Based Automatic Curriculum Learning in MuJoCo
SN - 978-989-758-737-5
AU - Chang C.
AU - Li C.
AU - Seyler J.
AU - Eivazi S.
PY - 2025
SP - 133
EP - 143
DO - 10.5220/0013116400003890
PB - SciTePress