Hacker News
- Lessons Learned Reproducing a Deep Reinforcement Learning Paper (2018) http://amid.fish/reproducing-deep-rl 11 comments
- Lessons Learned Reproducing a Deep Reinforcement Learning Paper http://amid.fish/reproducing-deep-rl 22 comments
- Show HN: Karplus-Strong Guitar Synthesizer in JavaScript http://amid.fish/javascript-karplus-strong/ 32 comments
- "Lessons Learned Reproducing a Deep Reinforcement Learning Paper" over 8 months: DRL is hard; sweat the small stuff; keep records of changes; measure everything; use cloud to iterate faster; run repeatedly due to instability (Amid Fish) http://amid.fish/reproducing-deep-rl 3 comments reinforcementlearning