android_env is a reinforcement learning (RL) environment developed by Google DeepMind that enables agents to interact with Android applications directly as a learning environment. It provides a standardized API for training agents to perform tasks on Android apps, supporting tasks ranging from games to productivity apps, making it suitable for research in real-world RL settings.

Features

  • Connects Android apps to RL agents through a standardized environment
  • Provides pixel observations and touch interactions for agents
  • Offers integration with RL libraries via Python API
  • Runs apps in Android Emulator to simulate real-world device behavior
  • Supports flexible reward specification based on app states or events
  • Enables research in human-computer interaction, accessibility, and automation

Project Samples

Project Activity

See All Activity >