Go Back
You can't go any further
You can't go any further
meritocratic
regular
democratic
hot
top
alive
4 posts
The Pointers Problem
59 posts
Value Learning
20
Stable Pointers to Value: An Agent Embedded in Its Own Utility Function
abramdemski
5y
9
93
The Pointers Problem: Human Values Are A Function Of Humans' Latent Variables
johnswentworth
2y
43
19
Stable Pointers to Value III: Recursive Quantilization
abramdemski
4y
4
23
Stable Pointers to Value II: Environmental Goals
abramdemski
4y
2
21
Latent Variables and Model Mis-Specification
jsteinhardt
4y
7
24
AI Alignment Problem: “Human Values” don’t Actually Exist
avturchin
3y
29
50
The easy goal inference problem is still hard
paulfchristiano
4y
19
51
Humans can be assigned any values whatsoever…
Stuart_Armstrong
4y
26
42
Since figuring out human values is hard, what about, say, monkey values?
shminux
2y
13
9
AIs should learn human preferences, not biases
Stuart_Armstrong
8mo
1
36
Human-AI Interaction
Rohin Shah
3y
10
13
An Open Philanthropy grant proposal: Causal representation learning of human preferences
PabloAMC
11mo
6
52
What is ambitious value learning?
Rohin Shah
4y
28
14
Can few-shot learning teach AI right from wrong?
Charlie Steiner
4y
3
20
Morally underdefined situations can be deadly
Stuart_Armstrong
1y
8
27
Learning human preferences: black-box, white-box, and structured white-box access
Stuart_Armstrong
2y
9
16
Values, Valence, and Alignment
Gordon Seidoh Worley
3y
4
9
What's the dream for giving natural language commands to AI?
Charlie Steiner
3y
8