Curiosity in AI
3,344 views · Published 2 November 2018 · 6:17 · Indexed 26 September 2026
Channel: CodeEmporium · 2018 · Education
Reinforcement learning generally uses a carrot-and-stick approach. Good actions are rewarded and bad actions are punished. But what are the drawbacks of this simple approach? How can we use curiosity to overcome it? Let's find out in this video! REFERENCES [1] Main Paper (Episodic Curiosity through Reachability) : https://arxiv.org/pdf/1810.02274.pdf [2] Blog: https://ai.googleblog.com/2018/10/curiosity-and-procrastination-in.html [3] AI learns to walk: https://deepmind.com/blog/producing-flexible-behaviours-simulated-environments/
