My thoughts on What rebuilding Alpha Go teaches us about AI andR RL: What rebuilding Alpha Go teaches us about AI andR RL
Commentary
- Dwarkesh Podcast: What rebuilding Alpha Go teaches us about AI andR RL
- This is such a great video. It explains or I should say converses in such a understandable human language, rather than buzzwords and in the air topics.
- I haven’t completed watching the full video, but so far it has given me a great understanding of what Go actually is, what problems it has compared to chess, and also how AI is being used to self-play it.
- I like Go now as well, the reason I love chess is its deterministic yet infinite possibilities, there is no luck factor. The mathematics required to solve such complex games is very interesting, a bit confusing at first, but it can be understandable if put enough thinking and time in it.