Go Back
You can't go any further
You can't go any further
meritocratic
regular
democratic
hot
top
alive
6 posts
Reward Functions
10 posts
Wireheading
72
Seriously, what goes wrong with "reward the agent when it makes you smile"?
TurnTrout
4mo
41
89
Scaling Laws for Reward Model Overoptimization
leogao
2mo
11
23
The reward engineering problem
paulfchristiano
3y
3
43
A Short Dialogue on the Meaning of Reward Functions
Leon Lang
1mo
0
13
$100/$50 rewards for good references
Stuart_Armstrong
1y
5
44
Reward model hacking as a challenge for reward learning
Erik Jenner
8mo
1
42
Four usages of "loss" in AI
TurnTrout
2mo
18
61
Towards deconfusing wireheading and reward maximization
leogao
3mo
7
16
Wireheading as a potential problem with the new impact measure
Stuart_Armstrong
4y
20
17
Wireheading is in the eye of the beholder
Stuart_Armstrong
3y
10
12
Model-based RL, Desires, Brains, Wireheading
Steven Byrnes
1y
1
19
Wireheading and discontinuity
Michele Campolo
2y
4
41
Draft papers for REALab and Decoupled Approval on tampering
Jonathan Uesato
2y
2
7
Note on algorithms with multiple trained components
Steven Byrnes
7h
1
9
Value extrapolation vs Wireheading
Stuart_Armstrong
6mo
1
15
Defining AI wireheading
Stuart_Armstrong
3y
9