Go Back
Choose this branch
Choose this branch
meritocratic
regular
democratic
hot
top
alive
4 posts
Human Values
Heuristics & Biases
Aesthetics
16 posts
Gradient Hacking
Evolution
Information Theory
Modularity
Gradient Descent
Biology
Experiments
Request Post
Cultural knowledge
92
Human values & biases are inaccessible to the genome
TurnTrout
5mo
51
117
A broad basin of attraction around human values?
Wei_Dai
8mo
16
11
Brain-over-body biases, and the embodied value problem in AI alignment
geoffreymiller
2mo
6
26
Preference synthesis illustrated: Star Wars
Stuart_Armstrong
2y
8
55
Gradient Hacker Design Principles From Biology
johnswentworth
3mo
13
49
Ten experiments in modularity, which we'd like you to run!
TheMcDouglas
6mo
2
170
Utility Maximization = Description Length Minimization
johnswentworth
1y
40
31
Gradient hacking: definitions and examples
Richard_Ngo
5mo
1
91
The Telephone Theorem: Information At A Distance Is Mediated By Deterministic Constraints
johnswentworth
1y
21
37
Conditions for mathematical equivalence of Stochastic Gradient Descent and Natural Selection
Oliver Sourbut
7mo
12
36
Theories of Modularity in the Biological Literature
TheMcDouglas
8mo
13
70
Gradient descent is not just more efficient genetic algorithms
leogao
1y
14
155
Evolution of Modularity
johnswentworth
3y
12
31
Hypothesis: gradient descent prefers general circuits
Quintin Pope
10mo
26
48
The Blackwell order as a formalization of knowledge
Alex Flint
1y
10
41
Emergent modularity and safety
Richard_Ngo
1y
15
11
Anthropic Effects in Estimating Evolution Difficulty
Mark Xu
1y
2
21
How to Throw Away Information
johnswentworth
3y
5