x

Clairstan

Message

1

3y

Optimization, loss set at variance in RL

This is a suggestion of a reinforcement learning model with an additional conflicting dynamic between optimized data and loss function. That conflict is intended to reduce RL’s intrinsic Omohundro x-risk dynamic. Thus, it’s also an attempt to produce, though without details, a good idea for AI safety, and so as...

Jul 22, 20231

Clairstan

Subscribe

Message

1

3y

Clairstan

Optimization, loss set at variance in RL

This is a suggestion of a reinforcement learning model with an additional conflicting dynamic between optimized data and loss function. That conflict is intended to reduce RL’s intrinsic Omohundro x-risk dynamic. Thus, it’s also an attempt to produce, though without details, a good idea for AI safety, and so as...

Jul 22, 20231

Optimization, loss set at variance in RL

Clairstan

3y

This is a suggestion of a reinforcement learning model with an additional conflicting dynamic between optimized data and loss function. That conflict is intended to reduce RL’s intrinsic Omohundro x-risk dynamic. Thus, it’s also an attempt to produce, though without details, a good idea for AI safety, and so as a test for the meta-usefulness of my posting here. That’s following the idea.

The Idea

A modification of reinforcement learning: to have the learning agent develop de sui a function for optimization which is to supplant the loss function, at a given reduction of the loss function, by the action of the learning process. That is, if we correlate the loss function not with arbitrary data to... (read 665 more words →)

1

LESSWRONG
LW

LESSWRONG
LW

Clairstan

Clairstan

Clairstan

Optimization, loss set at variance in RL

Clairstan

Clairstan

Clairstan

Optimization, loss set at variance in RL

The Idea