Reinforcement learning
A research field
Reinforcement learning is losing ground in AI research: 4.7% to 4.1% of new AI papers, -0.6 points. It ranks #44 of 44 by gain in share. As of Oct 10, 2026.
- Change in share
- -0.6 pts
- Share, last 3 months
- 4.1%
- Papers, last 3 months
- 1,367
- Ready for products
- 2.2 / 5
Ideas in this field
Credit assignment0.3% → 0.4%+0.0 ptsCurriculum reinforcement learning0.1% → 0.1%+0.0 ptsGroup relative policy optimization0.1% → 0.2%+0.0 ptsModel based reinforcement learning0.1% → 0.2%+0.0 ptsHierarchical reinforcement learning0.1% → 0.1%+0.0 ptsConstrained markov decision processe0.0% → 0.0%0.0 ptsProximal policy optimization0.0% → 0.0%0.0 ptsPrioritized experience replay0.0% → 0.0%0.0 ptsModel based planning0.0% → 0.0%0.0 ptsAction chunking0.0% → 0.0%0.0 pts
Companies working on it
Reinforcement learning: quick answers
No. Its share of new AI papers went from 4.7% to 4.1%, with 1,367 papers in the last 3 months.