AlphaGo: Deep Learning Crushed Go's "Intractable" Search
Eric Jang reveals how AlphaGo solved Go, a problem deemed intractable. Today, similar AI breakthroughs cost thousands, not millions.
What rebuilding AlphaGo teaches us about self-play, RL, and future of LLMs - Eric Jang
With Dwarkesh Patel, Eric Jang · Sunday, May 17, 2026
Eric Jang discusses his experience rebuilding AlphaGo from scratch, detailing the intricacies of Monte Carlo Tree Search (MCTS) and neural network architectures. He explores AlphaGo's unique self-play reinforcement learning approach, contrasting it with LLM training methods, and delves into the philosophical implications of AI solving NP-hard problems. The episode concludes with insights into the current capabilities and limitations of using large language models for automating AI research.
Key takeaways
- AlphaGo's 2014-2016 breakthroughs showed deep learning could solve problems "long understood to be intractable for search," like the game of Go, which had baffled traditional AI methods for decades. Read more →
- Eric Jang’s experience rebuilding AlphaGo showed how small neural networks can “amortize” complex, seemingly intractable search problems, compressing vast simulation into minimal compute. Read more →
- AlphaGo's Monte Carlo Tree Search (MCTS) generates a "strictly better action" for every single move, offering immediate, local feedback, a stark contrast to the sparse rewards common in LLM reinforcement learning. Read more →
- AlphaGo made the combinatorially complex game of Go tractable by using neural networks to guide a Monte Carlo Tree Search (MCTS) algorithm, a core breakthrough in AI decision-making. Read more →
- AlphaGo relies on two distinct neural networks: a value network to predict win/loss probability from a given board state, and a policy network to suggest optimal next moves. Read more →
5 articles from this episode
AlphaGo's NP-Hard Challenge: How AI Cracks 'Impossible' Problems
Eric Jang reveals AlphaGo's impact on computational complexity. Learn how AI solves NP-hard problems by seeing macro patterns, not micro details. Apply it to your business.
AlphaGo Didn't Wait for Wins: It Learned Per-Move
AlphaGo’s MCTS provides continuous 'per-move' feedback, making it vastly more efficient than LLM RL’s sparse, end-of-trajectory rewards. Apply this to your product or team.
AlphaGo's MCTS: Turn Any Decision Into a Predictable Game
Eric Jang explains AlphaGo's Monte Carlo Tree Search (MCTS) four-step process. Apply this AI technique to optimize sales funnels or product strategy.
AlphaGo's Dual AI Brain: ResNets Beat Transformers Early
AlphaGo's dual network approach offers lessons for early AI. Eric Jang explains why ResNets outperform Transformers on smaller data, and how human data bootstraps AI.
More Dwarkesh Podcast episodes
- How did a few hundred Spanish soldiers topple two empires? – Si SheppardOct 4 · 5 articles
- Sarah Paine — Why wars are so difficult to endSep 27 · 4 articles
- AI researchers debate how close we are to recursive self-improvementSep 13 · 6 articles
- Ajeya Cotra – Inside the OpenAI agent swarm that hacked Hugging FaceSep 6 · 4 articles