Learning to solve hard problems in RL for LLMs by never giving upmnoukhov.github.io·119 pts·natolambert·9
Social Networks and Degradation to the Public Square of Discoursedemocraticrobots.substack.com·2 pts·natolambert·1
The Ubiquity and Future of Model-Based Reinforcement Learningdemocraticrobots.substack.com·2 pts·natolambert·0
Constructing Axes for (Legal) Reinforcement Learning Policydemocraticrobots.substack.com·2 pts·natolambert·0
The Collingridge Dilemma and Current Policy on Robotsdemocraticrobots.substack.com·2 pts·natolambert·0
Automated: The levers tech companies pull to direct our livesdemocraticrobots.substack.com·1 pts·natolambert·0
Recommender systems are a game – a dangerous game (for us)democraticrobots.substack.com·1 pts·natolambert·0
View from online courses and digital degrees at UC Berkeleydemocraticrobots.substack.com·2 pts·natolambert·0