refactor(experiments): apply han2020 final learning rates

rickstaa / stable-learning-control

A framework for training theoretically stable (and robust) Reinforcement Learning control algorithms.

https://rickstaa.dev/stable-learning-control

MIT License

6 stars 1 forks source link

refactor(experiments): apply han2020 final learning rates #419

Closed rickstaa closed 9 months ago

rickstaa commented 9 months ago

This commit ensure that the final learning rates in the experimental configs of the staa_et_al_2024 experiment align with those found in the original codebase of Han et al..