Intelligent Agent Foundations Forumsign up / log in
Indifference and compensatory rewards
discussion post by Stuart Armstrong 36 days ago | discuss

It’s occurred to me that there is a framework where we can see all “indifference” results as corrective rewards, both for the utility function change indifference and for the policy change indifference.

Imagine that the agent has reward \(R_0\) and is following policy \(\pi_0\), and we want to change it to having reward \(R_1\) and following policy \(\pi_1\).

Then the corrective reward we need to pay it, so that it doesn’t attempt to resist or cause that change, is simply the difference between the two expected values:

  • \(V(R_0|\pi_0)-V(R_1|\pi_1)\),

where \(V\) is the agent’s own valuation of the expected reward, conditional on the policy.

This explains why off-policy reward-based agents are already safely interruptible: since we change the policy, not the reward, \(R_0=R_1\). And since off-policy agents have value estimates that are indifferent to the policy followed, \(V(R_0|\pi_0)=V(R_1|\pi_1)\), and the compensatory rewards are zero.



NEW LINKS

NEW POSTS

NEW DISCUSSION POSTS

RECENT COMMENTS

I don't know which open
by Jessica Taylor on Some problems with making induction benign, and ap... | 0 likes

KWIK learning is definitely
by Vadim Kosoy on Some problems with making induction benign, and ap... | 0 likes

I should have said "reliably
by Patrick LaVictoire on HCH as a measure of manipulation | 0 likes

I think that one can argue
by Vadim Kosoy on Generalizing Foundations of Decision Theory | 0 likes

"Having a well-calibrated
by Jessica Taylor on HCH as a measure of manipulation | 0 likes

Re #2, I think this is an
by Patrick LaVictoire on HCH as a measure of manipulation | 0 likes

Re #1, an obvious set of
by Patrick LaVictoire on HCH as a measure of manipulation | 0 likes

There's the additional
by Patrick LaVictoire on HCH as a measure of manipulation | 0 likes

I agree it's not a complete
by David Krueger on An idea for creating safe AI | 0 likes

I spoke with Huw about this
by David Krueger on An idea for creating safe AI | 0 likes

Both of your conjectures are
by Alex Mennen on Generalizing Foundations of Decision Theory | 0 likes

I can think of two problems:
by Ryan Carey on HCH as a measure of manipulation | 0 likes

Question that I haven't seen
by Patrick LaVictoire on All the indifference designs | 0 likes

Agree that IRL doesn't solve
by Jessica Taylor on Some problems with making induction benign, and ap... | 0 likes

Designing an agent which is
by Vadim Kosoy on An idea for creating safe AI | 0 likes

RSS

Privacy & Terms