paper-with-me

Papers

Benchmarking Partial Observability in Reinforcement Learning with a Suite of Memory-Improvable Domains

2025-07-31 · Ruo Yu Tao, Kaicheng Guo, Cameron Allen, George Konidaris arxiv

Mitigating partial observability is a necessary but challenging task for general reinforcement learning algorithms. To improve an algorithm's ability to mitigate partial observability, researchers need comprehensive benchmarks to gauge progress. Most algorithms tackling partial observability are only evaluated on benchmarks with simple forms of state aliasing, such as feature masking and Gaussian noise. Such benchmarks do not represent the many forms of partial observability seen in real domains, like visual occlusion or unknown opponent intent. We argue that a partially observable benchmark should have two key properties. The first is coverage in its forms of partial observability, to ensure an algorithm's generalizability. The second is a large gap between the performance of a agents with more or less state information, all other factors roughly equal. This gap implies that an environment is memory improvable: where performance gains in a domain are from an algorithm's ability to cope with partial observability as opposed to other factors. We introduce best-practice guidelines for empirically benchmarking reinforcement learning under partial observability, as well as the open-source library POBAX: Partially Observable Benchmarks in JAX. We characterize the types of partial observability present in various environments and select representative environments for our benchmark. These environments include localization and mapping, visual control, games, and more. Additionally, we show that these tasks are all memory improvable and require hard-to-learn memory functions, providing a concrete signal for partial observability research. This framework includes recommended hyperparameters as well as algorithm implementations for fast, out-of-the-box evaluation, as well as highly performant environments implemented in JAX for GPU-scalable experimentation.

📄 PDF Abstract BibTeX arXiv:2508.00046

Code (0)

등록된 구현이 없습니다.

Tasks

General Reinforcement Learning

Similar Papers 제목 키워드 기반

POPGym: Benchmarking Partially Observable Reinforcement Learning

2023-03-03 · Steven Morad, Ryan Kortvelesy, Matteo Bettini, Stephan Liwicki 외

Real world applications of Reinforcement Learning (RL) are often partially observable, thus requiring memory. Despite this, partial observability is still largely ignored by contemporary RL benchmarks and libraries. We i…

BenchmarkingGPUPartially Observable Reinforcement Learningreinforcement-learning+4

Reinforcement Learning using Guided Observability

2021-04-22 · Stephan Weigand, Pascal Klink, Jan Peters, Joni Pajarinen

Due to recent breakthroughs, reinforcement learning (RL) has demonstrated impressive performance in challenging sequential decision-making problems. However, an open question is how to make RL cope with partial observabi…

Decision MakingMuJoCoOpenAI GymOpen-Ended Question Answering+4

Memory, Benchmark & Robots: A Benchmark for Solving Complex Tasks with Reinforcement Learning

2025-02-14 · Egor Cherepanov, Nikita Kachaev, Alexey K. Kovalev, Aleksandr I. Panov

Memory is crucial for enabling agents to tackle complex tasks with temporal and spatial dependencies. While many reinforcement learning (RL) algorithms incorporate memory, the field lacks a universal benchmark to assess …

Reinforcement Learning (RL)Skills Assessment

The act of remembering: a study in partially observable reinforcement learning

2020-10-05 · Rodrigo Toro Icarte, Richard Valenzano, Toryn Q. Klassen, Phillip Christoffersen 외

Reinforcement Learning (RL) agents typically learn memoryless policies---policies that only consider the last observation when selecting actions. Learning memoryless policies is efficient and optimal in fully observable …

Partially Observable Reinforcement Learningreinforcement-learningReinforcement Learning (RL)

Recurrent Deep Reinforcement Learning for Chemotherapy Control under Partial Observability

2026-05-04 · Firas Mohamed Elamine Kiram, Imane Youkana, Rachida Saouli, Gian Antonio Susto 외 arxiv

Chemotherapy dose optimization can be formulated as a dynamic treatment regime, requiring sequential decisions under uncertainty that must balance tumor suppression against toxicity. However, most reinforcement learning …

Reinforcement Learning