扩展经典游戏环境，支持连续动作控制，统一评测强化学习算法。
CALE: Continuous Arcade Learning Environment
- 基于Stella模拟器，新增连续动作支持，兼容多种强化学习算法。
- 首次在相同环境上对比连续控制与值函数类算法，提供基准结果。
- 适合研究连续控制、多算法比较的学者，推动通用评估框架发展。
我们提出连续街机学习环境（CALE），作为知名街机学习环境（ALE）的扩展。CALE采用相同的Atari 2600模拟器（Stella），但增加了对连续动作的支持。这使得连续控制代理（如PPO和SAC）与基于值的代理（如DQN和Rainbow）可在同一环境套件中进行基准测试与评估。我们提出了若干开放性问题与研究方向，并提供了使用Soft Actor-Critic的初步基线结果。CALE已作为ALE的一部分开源，地址为https://github.com/Farama-Foundation/Arcade-Learning-Environment。
原文摘要 · Abstract (English)
We introduce the Continuous Arcade Learning Environment (CALE), an extension of the well-known Arcade Learning Environment (ALE) [Bellemare et al., 2013]. The CALE uses the same underlying emulator of the Atari 2600 gaming system (Stella), but adds support for continuous actions. This enables the benchmarking and evaluation of continuous-control agents (such as PPO [Schulman et al., 2017] and SAC [Haarnoja et al., 2018]) and value-based agents (such as DQN [Mnih et al., 2015] and Rainbow [Hessel et al., 2018]) on the same environment suite. We provide a series of open questions and research directions that CALE enables, as well as initial baseline results using Soft Actor-Critic. CALE is available as part of the ALE athttps://github.com/Farama-Foundation/Arcade-Learning-Environment.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。