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
252
Reward is not the optimization target
TurnTrout
4mo
97
86
Scaling Laws for Reward Model Overoptimization
leogao
2mo
11
82
Jitters No Evidence of Stupidity in RL
1a3orn
1y
18
76
Seriously, what goes wrong with "reward the agent when it makes you smile"?
TurnTrout
4mo
41
69
Towards deconfusing wireheading and reward maximization
leogao
3mo
7
59
Big picture of phasic dopamine
Steven Byrnes
1y
18
59
My take on Michael Littman on "The HCI of HAI"
Alex Flint
1y
4
47
Draft papers for REALab and Decoupled Approval on tampering
Jonathan Uesato
2y
2
42
Four usages of "loss" in AI
TurnTrout
2mo
18
40
A Short Dialogue on the Meaning of Reward Functions
Leon Lang
1mo
0
32
Conditioning, Prompts, and Fine-Tuning
Adam Jermyn
4mo
9
26
The reward engineering problem
paulfchristiano
3y
3
26
Wireheading is in the eye of the beholder
Stuart_Armstrong
3y
10
26
Reinforcement learning with imperceptible rewards
Vanessa Kosoy
3y
1
273
EfficientZero: How It Works
1a3orn
1y
42
134
EfficientZero: human ALE sample-efficiency w/MuZero+self-supervised
gwern
1y
52
114
We have achieved Noob Gains in AI
phdead
7mo
21
94
Evaluations project @ ARC is hiring a researcher and a webdev/engineer
Beth Barnes
3mo
7
87
The alignment problem in different capability regimes
Buck
1y
12
77
OpenAI Solves (Some) Formal Math Olympiad Problems
Michaël Trazzi
10mo
26
74
Will we run out of ML data? Evidence from projecting dataset size trends
Pablo Villalobos
1mo
12
51
Misc. questions about EfficientZero
Daniel Kokotajlo
1y
17
35
It matters when the first sharp left turn happens
Adam Jermyn
2mo
9
34
Remaking EfficientZero (as best I can)
Hoagy
5mo
9
5
Epistemic Strategies of Safety-Capabilities Tradeoffs
adamShimi
1y
0