paper-with-me

홈 › Papers

LiMIIRL: Lightweight Multiple-Intent Inverse Reinforcement Learning

2021-06-03 · Aaron J. Snoswell, Surya P. N. Singh, Nan Ye

Multiple-Intent Inverse Reinforcement Learning (MI-IRL) seeks to find a reward function ensemble to rationalize demonstrations of different but unlabelled intents. Within the popular expectation maximization (EM) framework for learning probabilistic MI-IRL models, we present a warm-start strategy based on up-front clustering of the demonstrations in feature space. Our theoretical analysis shows that this warm-start solution produces a near-optimal reward ensemble, provided the behavior modes satisfy mild separation conditions. We also propose a MI-IRL performance metric that generalizes the popular Expected Value Difference measure to directly assesses learned rewards against the ground-truth reward ensemble. Our metric elegantly addresses the difficulty of pairing up learned and ground truth rewards via a min-cost flow formulation, and is efficiently computable. We also develop a MI-IRL benchmark problem that allows for more comprehensive algorithmic evaluations. On this problem, we find our MI-IRL warm-start strategy helps avoid poor quality local minima reward ensembles, resulting in a significant improvement in behavior clustering. Our extensive sensitivity analysis demonstrates that the quality of the learned reward ensembles is improved under various settings, including cases where our theoretical assumptions do not necessarily hold. Finally, we demonstrate the effectiveness of our methods by discovering distinct driving styles in a large real-world dataset of driver GPS trajectories.

📄 PDF Abstract BibTeX arXiv:2106.01777

Code (0)

등록된 구현이 없습니다.

Tasks

Clusteringreinforcement-learningReinforcement LearningReinforcement Learning (RL)

Methods 이 논문이 사용한 방법론

GPS Greedy Policy Search (GPS) is a simple algorithm that learns a policy for test-time data augmentation based on the predictive performance on a validation set. GPS starts with…

Similar Papers 제목 키워드 기반

Deep multi-intentional inverse reinforcement learning for cognitive multi-function radar inverse cognition

2024-08-16 · HanCong Feng, Kaili Jiang, Bin Tang

In recent years, radar systems have advanced significantly, offering environmental adaptation and multi-task capabilities. These developments pose new challenges for electronic intelligence (Elint) and electronic support…

reinforcement-learningReinforcement LearningTrajectory Clustering

CoMI-IRL: Contrastive Multi-Intention Inverse Reinforcement Learning

2026-02-07 · Antonio Mone, Frans A. Oliehoek, Luciano Cavalcante Siebert arxiv

Inverse Reinforcement Learning (IRL) seeks to infer reward functions from expert demonstrations. When demonstrations originate from multiple experts with different intentions, the problem is known as Multi-Intention IRL …

Reinforcement Learning

Multi-intention Inverse Q-learning for Interpretable Behavior Representation

2023-11-23 · Hao Zhu, Brice De La Crompe, Gabriel Kalweit, Artur Schneider 외

In advancing the understanding of natural decision-making processes, inverse reinforcement learning (IRL) methods have proven instrumental in reconstructing animal's intentions underlying complex behaviors. Given the rec…

Decision MakingQ-Learning

Probabilistic Recurrent Intention Switching Model

2026-05-26 · Wenyuan Sheng, Hao Zhu, Joschka Boedecker arxiv

Inverse reinforcement learning (IRL) recovers reward functions from observed behavior, yet traditional methods assume a single stationary reward that cannot capture goal switching within an episode. Recent multi-intentio…

Reinforcement Learning

Extrapolating Beyond Suboptimal Demonstrations via Inverse Reinforcement Learning from Observations

2019-04-12 · Daniel S. Brown, Wonjoon Goo, Prabhat Nagarajan, Scott Niekum

A critical flaw of existing inverse reinforcement learning (IRL) methods is their inability to significantly outperform the demonstrator. This is because IRL typically seeks a reward function that makes the demonstrator …

Deep Reinforcement LearningImitation LearningMuJoCoreinforcement-learning+2