Mechanism design: Creates rules ensuring rational agents achieve intended outcomes
Image: NASA / Dominic Hart, Public domain, via Wikimedia Commons
Mechanism design: Creates rules ensuring rational agents achieve intended outcomes
Prisoner's dilemma
Prisoner's dilemma illustrates how individual rationality can lead to collectively worse outcomes
Reasoning model
RLMs excel in logic, math, and programming tasks
Monad (functional programming)
Monads are a type constructor with two operations: return and bind
constitutional AI does: model critiques and revises its own outputs using principles
Constitutional AI critiques and revises outputs using principles
branch prediction does: guesses which way an if-statement will go to keep the pipeline full
Speculative execution: Predicts if-statement outcomes to keep the pipeline full
the revelation principle says: any mechanism can be converted to a truthful one
Revelation principle: any mechanism can be converted to a truthful one
One email a day: 5 concepts + the 5 stories that matter →
Swipe through 100 ML concepts daily
Open TickerNews