Go Back
Choose this branch
Choose this branch
meritocratic
regular
democratic
hot
top
alive
63 posts
Decision Theory
Utility Functions
Quantilization
Mild Optimization
Bounded Rationality
Coherence Arguments
Law-Thinking
Orthogonality Thesis
Coherent Extrapolated Volition
Indexical Information
20 posts
Goal-Directedness
Literature Reviews
59
Take 7: You should talk about "the human's utility function" less.
Charlie Steiner
12d
22
64
Notes on "Can you control the past"
So8res
2mo
40
82
The "Measuring Stick of Utility" Problem
johnswentworth
6mo
22
134
Can you control the past?
Joe Carlsmith
1y
93
28
Quantilizers and Generative Models
Adam Jermyn
5mo
5
102
Coherence arguments imply a force for goal-directed behavior
KatjaGrace
1y
27
126
An Orthodox Case Against Utility Functions
abramdemski
2y
53
72
My Current Take on Counterfactuals
abramdemski
1y
57
28
Inferring utility functions from locally non-transitive preferences
Jan
10mo
15
154
Realism about rationality
Richard_Ngo
4y
145
105
Utility ≠ Reward
vlad_m
3y
25
60
Dutch-Booking CDT: Revised Argument
abramdemski
2y
22
10
Exploring Mild Behaviour in Embedded Agents
Megan Kinniment
5mo
3
59
What does it mean to apply decision theory?
abramdemski
2y
5
93
wrapper-minds are the enemy
nostalgebraist
6mo
36
92
why assume AGIs will optimize for fixed goals?
nostalgebraist
6mo
52
50
Finding Goals in the World Model
Jeremy Gillen
4mo
8
153
2021 AI Alignment Literature Review and Charity Comparison
Larks
12mo
26
137
2020 AI Alignment Literature Review and Charity Comparison
Larks
1y
14
69
When Most VNM-Coherent Preference Orderings Have Convergent Instrumental Incentives
TurnTrout
1y
4
43
P₂B: Plan to P₂B Better
Ramana Kumar
1y
14
69
Literature Review on Goal-Directedness
adamShimi
1y
21
29
Behavioral Sufficient Statistics for Goal-Directedness
adamShimi
1y
12
71
AI safety without goal-directed behavior
Rohin Shah
3y
15
17
Goal-Directedness and Behavior, Redux
adamShimi
1y
4
25
Against the Backward Approach to Goal-Directedness
adamShimi
1y
6
57
Will humans build goal-directed agents?
Rohin Shah
3y
43
20
Focus: you are allowed to be bad at accomplishing your goals
adamShimi
2y
17