This is a link post.This is a YouTube playlist of recorded lectures on the learning-theoretic AI alignment agenda (LTA) I gave for my MATS scholars of the Winter 2024 cohort, edited by my beloved spouse @Marcus Ogren. I hope these will become a useful resource for anyone who wants to get up to speed on the LTA, complementary to the reading list. In the future, I might record more lectures to expand this list.
** Table of Contents**
Agents and AIXI Hidden rewards and the problem of privilege Compositionality Nonrealizability It's a trap! Traps, continued Traps and frequentist guarantees Game theory and learning theory Hidden rewards Algorithmic Descriptive Agency Measure (ADAM) General reinforcement learning Infra-Bayesianism Learnability Infra-Bandits Newcombian problems Ultradistributions and semi-environments Formalizing Newcombian problems Pseudocausality and a general formulation of Newcombian problems Decision rules and pseudocausality Instrumental reward functions Infra-Bayesian haggling, part 1 Infra-Bayesian haggling, part 2 Anytime [...]
First published: October 27th, 2024
Source: https://www.lesswrong.com/posts/NWKk2eQwfuGzRXusJ/video-lectures-on-the-learning-theoretic-agenda)
---
Narrated by TYPE III AUDIO).