Evolution as Backstop for Reinforcement Learning · Gwern.net

Markets/evolution as backstops/ground truths for reinforcement learning/optimization: on some connections between Coase’s theory of the firm/linear optimization/DRL/evolution/multicellular life/pain/Internet communities as multi-level optimization problems.

--- title: Evolution as Backstop for Reinforcement Learning created: 2018-12-06 description: "Markets/evolution as backstops/ground truths for reinforcement learning/optimization: on some connections between Coase's theory of the firm/linear optimization/DRL/evolution/multicellular life/pain/Internet communities as multi-level optimization problems." status: finished modified: 2021-07-04 confidence: possible importance: 7 css-extension: dropcaps-kanzlei ... > One defense of free markets notes the inability of non-market mechanisms to solve planning & optimization problems. This has difficulty

saved by

• Amir
• Aryan Naik
• Owen Wang
• Jirat C
• Timothy Kostolansky
• khalid kaime
• Cheikh Fiteni
• Darius Kianersi
• Jackson Mowatt Gok

related reading

• Why Tool AIs Want to Be Agent AIs · Gwern.netgwern.net
• Reward is not the optimization target — LessWronglesswrong.com
• A Visual Guide to Evolution Strategies | 大トロblog.otoro.net
• Reward Is Not the Optimization Targetturntrout.com
• A (Long) Peek into Reinforcement Learning | Lil'Loglilianweng.github.io
• Debugging Reinforcement Learning Systemsandyljones.com
• Reinforcement Learning for Knowledge Awareness – kalomaze's kalomazing blogkalomaze.bearblog.dev
• You can’t imitation-learn how to continual-learn — LessWronglesswrong.com
• Models Don't "Get Reward" — LessWronglesswrong.com
• Human-like Neural Nets by Catapulting · Gwern.netgwern.net
• Reinforcement Learning: Machine Learning Meets Control Theoryyoutube.com
• It Looks Like You’re Trying To Take Over The World · Gwern.net (reader mode)gwern.net