Spaces:
Sleeping
Sleeping
Commit History
-
Transform into interactive RL playground for all audiences
Daksh C Jain +1: Claude Sonnet 4.6- 4-tab UI: Welcome, Playground, Algorithm Race, How it Works - DFS maze generator (guaranteed solvable corridors) + open random style - Animated GIF replay of agent solving the maze step by step - Q-Learning, SARSA, Monte Carlo — all with plain English descriptions - Algorithm Race tab: head-to-head convergence charts + winner announcement - Q-value heatmap showing what the agent learned - Difficulty presets: Tiny 5x5 to XL 17x17 - No jargon: human labels for all controls, story-driven onboarding - Modular structure: maze/, agents/, viz/ Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>34aeb9a