CALE: Continuous Arcade Learning Environment
This provides a standardized benchmark for evaluating reinforcement learning agents with continuous actions on Atari games, addressing a gap for researchers in machine learning and AI, though it is incremental as it builds on an existing environment.
The paper introduces the Continuous Arcade Learning Environment (CALE), an extension of the Arcade Learning Environment that adds support for continuous actions, enabling benchmarking of both continuous-control and value-based agents on the same Atari 2600 game suite, with initial baseline results using Soft Actor-Critic.
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.