Choose a paper to build a graph:

Paper Suggestions for "Memory Gym: Partially Observable Challenges to Memory-Based Agents"

1. Memory Gym: Partially Observable Challenges to Memory-Based Agents

Authors: Marco Pleines, M. Pallasch, Frank Zimmer, Mike Preuss
Event: International Conference on Learning Representations 2023
Citations: 0
References: 0
Notes: Not enough citations and references to build a graph

2. Memory Gym: Partially Observable Challenges to Memory-Based Agents in Endless Episodes

Authors: Marco Pleines, M. Pallasch, F. Zimmer, M. Preuss
Source: arXiv.org 2023
Citations: 4
References: 60
Summary: An implementation driven by TrXL and Proximal Policy Optimization is contributed, which leverages TrXL as episodic memory using a sliding window approach and makes a remarkable resurgence, consistently outperforming TrXL by significant margins.

3. Memory Gym: Towards Endless Tasks to Benchmark Memory Capabilities of Agents

Authors: Marco Pleines, M. Pallasch, Frank Zimmer, Mike Preuss
Source: Journal of machine learning research 2023
Citations: 14
References: 51

Summary: Memory Gym presents a suite of 2D partially observable environments, namely Mortar Mayhem, Mystery Path, and Searing Spotlights, designed to benchmark memory capabilities in decision-making agents. These environments...
Website and Source Code

4. From Tokens to Latent States: Leveraging Pre-trained Language Models for Improving Partially Observable Reinforcement Learning

Authors: Meiju Li, Ruixiang Sun, Xin Li, Mingzhong Wang
Event: AAAI Conference on Artificial Intelligence 2026
Citations: 0
References: 0
Summary: This paper observes a compelling analogy: large language models (LLMs) autoregressively generate token probability distributions based on preceding context, mirroring how belief states are maintained and updated in POMDPs.

5. Stable Hadamard Memory: Revitalizing Memory-Augmented Agents for Reinforcement Learning

Authors: Hung Le, Kien Do, D. Nguyen, Sunil Gupta, S. Venkatesh
Event: International Conference on Learning Representations 2024
Citations: 8
References: 40
Summary: Effective decision-making in partially observable environments demands robust memory management. Despite their success in supervised learning, current deep-learning memory models struggle in reinforcement learning environments that are partially observable and long-term.

6. Uncertainty-Aware Stochastic Hybrid World Models with Neural Map Memory for Autonomous Navigation in Partially Observable Grid Environments

Authors: Arwa Saad, Amer Ibrahim, Shashi Gupta
Event: Computational Discovery and Intelligent Systems (CDIS) 2026
Citations: 0
References: 29
Summary: This paper introduces SHWM-NM (Stochastic Hybrid World Model with Uncertainty-Aware Neural Map Memory), a unified framework that enhances autonomous navigation capabilities.

7. A Memory-Based Graph Reinforcement Learning Method for Critical Load Restoration With Uncertainties of Distributed Energy Resource

Authors: Bangji Fan, Xinghua Liu, Gaoxi Xiao, Yan Xu, Xiang Yang, Peng Wang
Source: IEEE Transactions on Smart Grid 2025
Citations: 5
References: 42
Summary: We develop a memory-based graph reinforcement learning approach, designed to train the agent to acquire a critical load restoration strategy in a distribution network under uncertainties.

8. Learning what to memorize: Using intrinsic motivation to form useful memory in partially observable reinforcement learning

Authors: Alper Demir
Source: Applied intelligence (Boston) 2021
Citations: 5
References: 40
Summary: We formalize an intrinsic motivation to support this learning mechanism, which guides the agent to memorize distinctive events and enable it to disambiguate its state in the environment.

9. Synthetic POMDPs to Challenge Memory-Augmented RL: Memory Demand Structure Modeling

Authors: Yongyi Wang, Lingfeng Li, Bozhou Chen, Ang Li, Hanyu Liu, Qirui Zheng, Xionghui Yang, Wenxin Li
Source: arXiv.org 2025
Citations: 1
References: 36
Summary: This paper advances the design of customizable POMDPs with three key contributions...

10. Interpretable Navigation Agents Using Attention-Augmented Memory

Authors: Jia Qu, S. Miwa, Y. Domae
Event: IEEE International Conference on Systems, Man and Cybernetics 2022
Citations: 3
References: 28
Summary: We propose a low-computational-complexity and scalable DRL model that uses attention-augmented memory (AAM)...