Eliezer_Yudkowsky comments on Newcomb's Problem and Regret of Rationality - Less Wrong

64 Post author: Eliezer_Yudkowsky 31 January 2008 07:36PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (588)

Sort By: Old

You are viewing a single comment's thread.

Comment author: Eliezer_Yudkowsky 01 February 2008 04:15:19AM 8 points [-]

Hanson: I suppose I might still be missing something, but this still seems to me just a simple example of time inconsistency

In my motivations and in my decision theory, dynamic inconsistency is Always Wrong. Among other things, it always implies an agent unstable under reflection.

A more familiar example is that we'd like to commit ahead of time to spending effort to punish people who hurt us, but after they hurt us we'd rather avoid spending that effort as the harm is already done.

But a self-modifying agent would modify to not rather avoid it.

Gowder: If a), then practical reason is meaningless anyway: you'll do what you'll do, so stop stressing about it.

Deterministic != meaningless. Your action is determined by your motivations, and by your decision process, which may include your stressing about it. It makes perfect sense to say: "My future decision is determined, and my stressing about it is determined; but if-counterfactual I didn't stress about it, then-counterfactual my future decision would be different, so it makes perfect sense for me to stress about this, which is why I am deterministically doing it."

The past can't change - does not even have the illusion of potential change - but that doesn't mean that people who, in the past, committed a crime, are not held responsible just because their action and the crime are now "fixed". It works just the same way for the future. That is: a fixed future should present no more problem for theories of moral responsibility than a fixed past.