Syllabus
Syllabus copied to clipboard
Synchronized Curriculum Learning for RL Agents
With Ray and native multiprocessing support, multiple curriculum learning methods, and pufferlib, it should be possible to easily use Syllabus for any problem with little to no engineering. It may...
[Prioritized Level Replay (PLR)](https://arxiv.org/pdf/2010.03934.pdf) is a simple-yet effective baseline for curriculum learning methods, and has been used in many prominent works ([AdA](https://arxiv.org/abs/2301.07608, [MAESTRO](https://openreview.net/forum?id=sKWlRDzPfd7), etc). It would help adoption of Syllabus...
step_queue should not be a required argument because it is not used when update_on_step is disabled in the environment sync wrapper. Additionally, the environment sync wrapper should return None instead...