A Continual Offline Reinforcement Learning Benchmark For Navigation Tasks

Abstract

Autonomous agents operating in domains such as robotics or video game simulations must adapt to changing tasks without forgetting about the previous ones. This process called Continual Reinforcement Learning poses non-trivial difficulties, from preventing catastrophic forgetting to ensuring the scalability of the approaches considered. Building on recent advances, we introduce a benchmark providing a suite of video-game navigation scenarios, thus filling a gap in the literature and capturing key challenges : catastrophic forgetting, task adaptation, and memory efficiency. We define a set of various tasks and datasets, evaluation protocols, and metrics to assess the performance of algorithms, including state-of-the-art baselines. Our benchmark is designed not only to foster reproducible research and to accelerate progress in continual reinforcement learning for gaming, but also to provide a reproducible framework for production pipelines -- helping practitioners to identify and to apply

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keykobanda2025a

Related papers

• CORA: Benchmarks, Baselines, And Metrics As A Platform For Continual Reinforcement Learning Agents (2021)0.00
• Challenges And Opportunities In Offline Reinforcement Learning From Visual Observations (2022)0.00
• Benchmark Environments For Multitask Learning In Continuous Domains (2017)0.00
• Continual World: A Robotic Benchmark For Continual Reinforcement Learning (2021)0.00
• AD4RL: Autonomous Driving Benchmarks For Offline Reinforcement Learning With Value-based Dataset (2024)7.16
• Task-agnostic Continual Reinforcement Learning: Gaining Insights And Overcoming Challenges (2022)0.00
• Neorl-2: Near Real-world Benchmarks For Offline Reinforcement Learning With Extended Realistic Scenarios (2025)0.00
• A Benchmark Environment For Offline Reinforcement Learning In Racing Games (2024)2.26