Go Back
Choose this branch
Choose this branch
meritocratic
regular
democratic
hot
top
alive
30 posts
Reinforcement Learning
Wireheading
Reward Functions
11 posts
AI Capabilities
EfficientZero
Tradeoffs
233
Reward is not the optimization target
TurnTrout
4mo
97
87
Jitters No Evidence of Stupidity in RL
1a3orn
1y
18
83
Scaling Laws for Reward Model Overoptimization
leogao
2mo
11
80
Seriously, what goes wrong with "reward the agent when it makes you smile"?
TurnTrout
4mo
41
77
Towards deconfusing wireheading and reward maximization
leogao
3mo
7
66
My take on Michael Littman on "The HCI of HAI"
Alex Flint
1y
4
53
Draft papers for REALab and Decoupled Approval on tampering
Jonathan Uesato
2y
2
42
Four usages of "loss" in AI
TurnTrout
2mo
18
38
Conditioning, Prompts, and Fine-Tuning
Adam Jermyn
4mo
9
38
Big picture of phasic dopamine
Steven Byrnes
1y
18
37
A Short Dialogue on the Meaning of Reward Functions
Leon Lang
1mo
0
35
Wireheading is in the eye of the beholder
Stuart_Armstrong
3y
10
34
Wireheading as a potential problem with the new impact measure
Stuart_Armstrong
4y
20
29
The reward engineering problem
paulfchristiano
3y
3
212
EfficientZero: How It Works
1a3orn
1y
42
144
EfficientZero: human ALE sample-efficiency w/MuZero+self-supervised
gwern
1y
52
108
We have achieved Noob Gains in AI
phdead
7mo
21
98
The alignment problem in different capability regimes
Buck
1y
12
81
Evaluations project @ ARC is hiring a researcher and a webdev/engineer
Beth Barnes
3mo
7
77
OpenAI Solves (Some) Formal Math Olympiad Problems
Michaël Trazzi
10mo
26
70
Misc. questions about EfficientZero
Daniel Kokotajlo
1y
17
44
Will we run out of ML data? Evidence from projecting dataset size trends
Pablo Villalobos
1mo
12
38
It matters when the first sharp left turn happens
Adam Jermyn
2mo
9
25
Remaking EfficientZero (as best I can)
Hoagy
5mo
9
8
Epistemic Strategies of Safety-Capabilities Tradeoffs
adamShimi
1y
0