What question did this study set out to answer?

This research aims to develop a framework that addresses challenges in multi-task learning within inverse reinforcement learning.

February 19, 2026Open Access

Adversarial Distributed Multi-Task Meta-Inverse Reinforcement Learning with Theory of Mind and Mean-Field Method

Key Points

This research aims to develop a framework that addresses challenges in multi-task learning within inverse reinforcement learning.
Developed a distributed multi-task framework called TMMF-MTAIRL.
Utilized theory of mind to capture relationships among tasks.
Integrated mean-field theory to simplify task interactions.
Introduced latent variables for better adaptation to new tasks.
Evaluated performance using benchmarks and real-world datasets.
TMMF-MTAIRL outperformed baseline methods in all tasks.
Achieved an average improvement of 0.16 in fault classification accuracy.
Demonstrated enhanced adaptation to novel tasks across multi-task environments.

Abstract

Maximum entropy adversarial inverse reinforcement learning (ME-AIRL) has garnered widespread attention for its ability to learn rewards and optimize policies from expert demonstrations. In complex multi-task environments, applying meta-learning ME-AIRL to acquire rewards requires a substantial volume of homogeneous expert demonstrations across all tasks, which is often impractical in real-world scenarios. Moreover, interference between tasks further escalates computational complexity. To solve these challenges, this paper proposes a distributed multi-task meta ME-AIRL framework based on theory of mind and mean field, referred to as TMMF-MTAIRL. In TMMF-MTAIRL, the theory of mind is used to capture the relationships and representational information among multiple tasks. Furthermore, TMMF-MTAIRL integrates mean-field theory to transform interactions between complex tasks into interactions between the main task and the average of the remaining tasks. Furthermore, additional latent variables are introduced to enhance adaptation to novel tasks. We evaluate the proposed TMMF-MTAIRL on point-maze benchmarks and a real-world rolling bearing fault diagnosis dataset using metrics such as classification accuracy, mean rewards or cumulative rewards. TMMF-MTAIRL achieves the best performance across all tasks, with an average improvement of 0.16 in accuracy of fault classification over the strongest baseline.

Demander à l'IA

Bookmark

View Full Paper