paper-with-me

홈 › Papers

Multistep Quasimetric Learning for Scalable Goal-conditioned Reinforcement Learning

2025-11-11 · Bill Chunyuan Zheng, Vivek Myers, Benjamin Eysenbach, Sergey Levine arxiv

Learning how to reach goals in an environment is a longstanding challenge in AI, yet reasoning over long horizons remains a challenge for modern methods. The key question is how to estimate the temporal distance between pairs of observations. While temporal difference methods leverage local updates to provide optimality guarantees, they often perform worse than Monte Carlo methods that perform global updates (e.g., with multi-step returns), which lack such guarantees. We show how these approaches can be integrated into a practical offline GCRL method that fits a quasimetric distance using a multistep Monte-Carlo return. We show our method outperforms existing offline GCRL methods on long-horizon simulated tasks with up to 4000 steps, even with visual observations. We also demonstrate that our method can enable stitching in the real-world robotic manipulation domain (Bridge setup). Our approach is the first end-to-end offline GCRL method that enables multistep stitching in this real-world manipulation domain from an unlabeled offline dataset of visual observations and demonstrate robust horizon generalization.

📄 PDF Abstract BibTeX arXiv:2511.07730

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Goal Reaching with Eikonal-Constrained Hierarchical Quasimetric Reinforcement Learning

2025-12-12 · Vittorio Giammarino, Ahmed H. Qureshi arxiv

Goal-Conditioned Reinforcement Learning (GCRL) mitigates the difficulty of reward design by framing tasks as goal reaching rather than maximizing hand-crafted reward signals. In this setting, the optimal goal-conditioned…

Reinforcement Learning

Offline Goal-conditioned Reinforcement Learning with Quasimetric Representations

2025-09-24 · Vivek Myers, Bill Chunyuan Zheng, Benjamin Eysenbach, Sergey Levine arxiv

Approaches for goal-conditioned reinforcement learning (GCRL) often use learned state representations to extract goal-reaching policies. Two frameworks for representation structure have yielded particularly effective GCR…

Representation LearningReinforcement LearningContrastive Learning

Offline Goal-Conditioned Reinforcement Learning with Projective Quasimetric Planning

2025-06-23 · Anthony Kobanda, Waris Radji, Mathieu Petitbois, Odalric-Ambrym Maillard 외

Offline Goal-Conditioned Reinforcement Learning seeks to train agents to reach specified goals from previously collected trajectories. Scaling that promises to long-horizon tasks remains challenging, notably due to compo…

Metric Learningreinforcement-learningReinforcement Learning

Intrinsic-Energy Joint Embedding Predictive Architectures Induce Quasimetric Spaces

2026-02-12 · Anthony Kobanda, Waris Radji arxiv

Joint-Embedding Predictive Architectures (JEPAs) aim to learn representations by predicting target embeddings from context embeddings, inducing a scalar compatibility energy in a latent space. In contrast, Quasimetric Re…

Reinforcement Learning

Optimal Goal-Reaching Reinforcement Learning via Quasimetric Learning

2023-04-03 · Tongzhou Wang, Antonio Torralba, Phillip Isola, Amy Zhang

In goal-reaching reinforcement learning (RL), the optimal value function has a particular geometry, called quasimetric structure. This paper introduces Quasimetric Reinforcement Learning (QRL), a new RL method that utili…

reinforcement-learningReinforcement LearningReinforcement Learning (RL)