AI ALIGNMENT FORUMTags
AF

Reward Functions

EditHistorySubscribe

Help improve this page (2 flags)

EditHistorySubscribe

Help improve this page (2 flags)

Reward Functions

Contributors

Posts tagged Reward Functions

3

95Reward is not the optimization target

2y

88

3

28Draft papers for REALab and Decoupled Approval on tampering

Jonathan Uesato, Ramana Kumar

4y

2

0

53Scaling Laws for Reward Model Overoptimization

leogao, John Schulman, Jacob Hilton

2y

5

2

40Seriously, what goes wrong with "reward the agent when it makes you smile"?

Alex Turner, johnswentworth

2y

13

1

39Interpreting Preference Models w/ Sparse Autoencoders

Logan Riggs Smith, Jannik Brinkmann

2mo

10

2

20Four usages of "loss" in AI

2y

16

0

15Language Agents Reduce the Risk of Existential Catastrophe

Cameron Domenico Kirk-Giannini, Simon Goldstein

1y

1

1

11$100/$50 rewards for good references

Stuart Armstrong

3y

2

1

55Utility ≠ Reward

Vladimir Mikulik

5y

16

0

41Reward hacking behavior can generalize across tasks

Kei, Isaac Dunn, Henry Sleight, Miles Turpin, Evan Hubinger, Carson Denison, Ethan Perez

4mo

1

0

13Shutdown-Seeking AI

Simon Goldstein

1y

5

1

27A Short Dialogue on the Meaning of Reward Functions

Leon Lang, Quintin Pope, peligrietzer

2y

0

1

10Thoughts on reward engineering

Paul Christiano

6y

19

1

6The reward engineering problem

Paul Christiano

6y

1

0

0Reward model hacking as a challenge for reward learning

2y

0