3 Hits in 1.1 sec

AndroidEnv: A Reinforcement Learning Platform for Android [article]

Daniel Toyama, Philippe Hamel, Anita Gergely, Gheorghe Comanici, Amelia Glaese, Zafarali Ahmed, Tyler Jackson, Shibl Mourad, Doina Precup
2021 arXiv   pre-print
We introduce AndroidEnv, an open-source platform for Reinforcement Learning (RL) research built on top of the Android ecosystem.  ...  on a set of tasks built on this platform.  ...  Acknowledgements It would have been impossible to create AndroidEnv without the help of many people: • Mathieu Méa for integrating dozens of open source apps.  ... 
arXiv:2105.13231v1 fatcat:6z24svakpnbw3mchlwctbbhylm

Learning how to Interact with a Complex Interface using Hierarchical Reinforcement Learning [article]

Gheorghe Comanici, Amelia Glaese, Anita Gergely, Daniel Toyama, Zafarali Ahmed, Tyler Jackson, Philippe Hamel, Doina Precup
2022 arXiv   pre-print
Hierarchical Reinforcement Learning (HRL) allows interactive agents to decompose complex problems into a hierarchy of sub-tasks.  ...  In this work, we study the utility of hierarchical decompositions for learning an appropriate way to interact with a complex interface.  ...  Hamel, A. Gergely, G. Comanici, A. Glaese, Z. Ahmed, T. Jackson, S. Mourad, and D. Precup. Androidenv: A reinforcement learning platform for android, 2021. A. White.  ... 
arXiv:2204.10374v1 fatcat:iz4olhl5vrhethepxqqdxrgipu

Learning UI Navigation through Demonstrations composed of Macro Actions [article]

Wei Li
2021 arXiv   pre-print
The action space is restricted to the UI elements plus a few global actions. Actions can be customized for tasks and each action is a sequence of basic operations conditioned on status checks.  ...  Demos are only collected for the failed cases during the evaluation of the previous version of the agent.  ...  Acknowledgements We would thank Daniel Rodriguez, Chris Rawles, and Alice Li for periodical discussions and proof reading of the write up.  ... 
arXiv:2110.08653v1 fatcat:kz7iihtzkvbu3pbaktujmxe5ia