Hacker News
- Solving Crew Battle Strategy with Math https://www.alexirpan.com/2024/03/23/crew-battle.html 22 comments
- My AI Timelines Have Sped Up (Again) https://www.alexirpan.com/2024/01/10/ai-timelines-2024.html 95 comments
- My AI Timelines Have Sped Up https://www.alexirpan.com/2020/08/18/ai-timelines.html 130 comments
- Deep reinforcement learning doesn't work yet https://www.alexirpan.com/2018/02/14/rl-hard.html 50 comments
- Deep Reinforcement Learning Doesn't Work (yet) https://www.alexirpan.com/2018/02/14/rl-hard.html 2 comments
- A Gentle Introduction to Secure Computation http://www.alexirpan.com/2016/02/11/secure-computation.html 6 comments
- Deep Reinforcement Learning Doesn't Work Yet (Feb 2018) https://www.alexirpan.com/2018/02/14/rl-hard.html 3 comments reinforcementlearning
- "Deep Reinforcement Learning Doesn't Work Yet": sample-inefficient, outperformed by domain-specific models or techniques, fragile reward functions, gets stuck in local optima, unreproducible & undebuggable, & doesn't generalize https://www.alexirpan.com/2018/02/14/rl-hard.html 9 comments reinforcementlearning