4 ms·
The next step seems to be applying past advances in reinforcement learning with modern transformer based models
by thisismyswamp 3y ago
The next step seems to be applying past advances in reinforcement learning with modern transformer based models
- mattsan 3y agoWhich multiple teams are working on - OpenAI (Q*), and Meta just released a reinforcement learning framework
- npsomaratna 3y agoCould you point me towards Meta's reinforcement learning framework? I'd like to see how it stacks up against the OpenAI gym.
- mattsan 3y agoSure thing - https://pearlagent.github.io/ https://pearlagent.github.io/ HN post here: https://news.ycombinator.com/item?id=38564526 https://news.ycombinator.com/item?id=38564526
- npsomaratna 3y agoThank you!