Intelligent Agent Foundations Forumsign up / log in
by Jessica Taylor 924 days ago | link | parent

Nice. I think this maps more precisely in the case where you ask questions about yourself. In this case, you can (in some cases) be assured that you will find no proof about what action you will take, but you can still prove things about logical consequences of your action. So the difference between the “shorter” proof and the “longer” proof is that you can’t actually find the longer proof.

My intuition about logical counterfactuals in general is that it seems easier to just solve agent-simulates-predictor type problems directly than to solve logical counterfactuals, since at least there’s a kind of win condition for ASP-type problems. So the statements about proof length here might map to facts about whether a bounded predictor can predict things about you. In ASP we want “Bounded predictor predicts that agent do X” to be considered a logical counterfactual of “agent does X” from the agent’s perspective; this would require there to be a short proof of “agent does X -> bounded predictor predicts agent does X”. But we don’t get this automatically: this would require the agent to use a reasoning system that the predictor can reason about somehow. I might write another post on intuitions about ASP-type problems.



NEW LINKS

NEW POSTS

NEW DISCUSSION POSTS

RECENT COMMENTS

Indeed there is some kind of
by Vadim Kosoy on Catastrophe Mitigation Using DRL | 0 likes

Very nice. I wonder whether
by Vadim Kosoy on Hyperreal Brouwer | 0 likes

Freezing the reward seems
by Vadim Kosoy on Resolving human inconsistency in a simple model | 0 likes

Unfortunately, it's not just
by Vadim Kosoy on Catastrophe Mitigation Using DRL | 0 likes

>We can solve the problem in
by Wei Dai on The Happy Dance Problem | 1 like

Maybe it's just my browser,
by Gordon Worley III on Catastrophe Mitigation Using DRL | 2 likes

At present, I think the main
by Abram Demski on Looking for Recommendations RE UDT vs. bounded com... | 0 likes

In the first round I'm
by Paul Christiano on Funding opportunity for AI alignment research | 0 likes

Fine with it being shared
by Paul Christiano on Funding opportunity for AI alignment research | 0 likes

I think the point I was
by Abram Demski on Predictable Exploration | 0 likes

(also x-posted from
by Sören Mindermann on The Three Levels of Goodhart's Curse | 0 likes

(x-posted from Arbital ==>
by Sören Mindermann on The Three Levels of Goodhart's Curse | 0 likes

>If the other players can see
by Stuart Armstrong on Predictable Exploration | 0 likes

Thinking about this more, I
by Abram Demski on Predictable Exploration | 0 likes

> So I wound up with
by Abram Demski on Predictable Exploration | 0 likes

RSS

Privacy & Terms