Reinforcement learning: learning from reward
No labels, no right answers. Only +1 for reaching the goal and −1 for falling in the pit. The agent solves it on its own in 400 episodes.
4 steps
175 XP
A free account is needed
Start the lesson →
Sources
- Sutton, R. S. & Barto, A. G. 2018 · Reinforcement Learning: An Introduction, 2. baskı, Bölüm 6.5 · MIT Press
- Watkins, C. J. C. H. & Dayan, P. 1992 · Q-learning · Machine Learning, 8(3-4)
- Mnih, V. et al. 2015 · Human-level Control through Deep Reinforcement Learning · Nature, 518
ML Academy · an interactive machine learning course that runs in your browser ·
All lessons