Intelligent Agent Foundations Forumsign up / log in
Index of some decision theory posts
discussion post by Tsvi Benson-Tilsen 136 days ago | Ryan Carey, Jack Gallagher, Jessica Taylor and Scott Garrabrant like this | discuss

What this is

An index of posts outlining some ideas in decision theory. I plan to have them available both on the forum and on github as pdfs. I might turn this into a general index for agent-foundations-related decision theory research depending on interest and my time.

Index

Index of some decision theory posts

As advertised.

Forum post: https://agentfoundations.org/item?id=1026

Github pdf: https://github.com/tsvibt/public-pdfs/blob/master/decision-theory/index/main.pdf

Notation for induction and decision theory

A reference for notation that might be useful for using (universal) Garrabrant inductors as models for bounded reasoning, and some notation for modelling agents. (Not on the forum because tables.)

Github pdf: https://github.com/tsvibt/public-pdfs/blob/master/decision-theory/notation/main.pdf

Desiderata for decision theory

A list of desiderata for a theory of optimal decision-making for bounded rational agents in general environments.

Forum post: https://agentfoundations.org/item?id=1053

Github pdf: https://github.com/tsvibt/public-pdfs/blob/master/decision-theory/desiderata/main.pdf

An inductive setting for decision theory

A discussion of the appropriate setting for studying decision theory.

Forum post: todo

Github pdf: todo

Training a universal Garrabrant inductor to predict counterfactuals

A proposal for training UGIs to predict action- and policy-counterfactuals by learning from the consequences of actions taken by similar (“logically previous”) agents.

Forum post: https://agentfoundations.org/item?id=1054

Github pdf: https://github.com/tsvibt/public-pdfs/blob/master/decision-theory/training-counterfactuals/main.pdf

Open problem: very thin logical priors

An open problem relevant to decision theory and to understanding bounded reasoning: is there a very easily computable prior over logical facts that, when updated on the results of computations, performs well in some sense?

Forum post: todo

Github pdf: todo



NEW LINKS

NEW POSTS

NEW DISCUSSION POSTS

RECENT COMMENTS

Why wouldn't it work? The
by Jessica Taylor on True understanding comes from passing exams | 0 likes

It would be weird if the
by Jessica Taylor on Are daemons a problem for ideal agents? | 0 likes

The second AI doesn't get to
by Stuart Armstrong on True understanding comes from passing exams | 0 likes

Fixed the $\varepsilon$,
by Scott Garrabrant on Entangled Equilibria and the Twin Prisoners' Dilem... | 0 likes

I think you meant to divide
by Vadim Kosoy on Entangled Equilibria and the Twin Prisoners' Dilem... | 0 likes

Yup, this isn't robust to
by Patrick LaVictoire on Censoring out-of-domain representations | 0 likes

I don't think "honesty" is
by Paul Christiano on How likely is a random AGI to be honest? | 2 likes

Discussed briefly in Concrete
by Daniel Dewey on Minimizing Empowerment for Safety | 2 likes

I reason as follows: 1.
by David Krueger on Does UDT *really* get counter-factually mugged? | 1 like

I agree... if there are
by David Krueger on Censoring out-of-domain representations | 0 likes

Game-aligned agents aren't
by Vladimir Nesov on Does UDT *really* get counter-factually mugged? | 0 likes

The issue in the OP is that
by Vladimir Nesov on Does UDT *really* get counter-factually mugged? | 0 likes

This seems only loosely
by David Krueger on Does UDT *really* get counter-factually mugged? | 0 likes

OK that makes sense, thanks.
by David Krueger on Does UDT *really* get counter-factually mugged? | 0 likes

It's not the same (but
by David Krueger on Learning Impact in RL | 0 likes

RSS

Privacy & Terms (NEW 04/01/15)