Go Back
Choose this branch
Choose this branch
meritocratic
regular
democratic
hot
top
alive
45 posts
Instrumental Convergence
Orthogonality Thesis
12 posts
Deconfusion
Gradient Hacking
Gradient Descent
59
You can still fetch the coffee today if you're dead tomorrow
davidad
11d
15
34
Empowerment is (almost) All We Need
jacob_cannell
1mo
43
0
The Opportunity and Risks of Learning Human Values In-Context
Zachary Robertson
10d
4
14
Is the Orthogonality Thesis true for humans?
Noosphere89
1mo
18
85
Instrumental convergence is what makes general intelligence possible
tailcalled
1mo
11
39
A caveat to the Orthogonality Thesis
Wuschel Schulz
1mo
10
7
[ASoT] Instrumental convergence is useful
Ulisse Mini
1mo
9
17
Misalignment-by-default in multi-agent systems
Edouard Harris
2mo
8
194
Seeking Power is Often Convergently Instrumental in MDPs
TurnTrout
3y
38
32
Instrumental convergence in single-agent systems
Edouard Harris
2mo
4
78
Coherence arguments imply a force for goal-directed behavior
KatjaGrace
1y
27
239
Debate on Instrumental Convergence between LeCun, Russell, Bengio, Zador, and More
Ben Pace
3y
60
11
Against Instrumental Convergence
zulupineapple
4y
31
3
Amending the "General Pupose Intelligence: Arguing the Orthogonality Thesis"
diegocaleiro
9y
22
27
Applications for Deconfusing Goal-Directedness
adamShimi
1y
3
12
Why Do AI researchers Rate the Probability of Doom So Low?
Aorou
2mo
6
14
Some real examples of gradient hacking
Oliver Sourbut
1y
8
17
Approaches to gradient hacking
adamShimi
1y
8
14
Gradient hacking: definitions and examples
Richard_Ngo
5mo
1
69
Looking Deeper at Deconfusion
adamShimi
1y
13
51
Hypothesis: gradient descent prefers general circuits
Quintin Pope
10mo
26
6
(Extremely) Naive Gradient Hacking Doesn't Work
ojorgensen
9h
0
40
Gradient descent is not just more efficient genetic algorithms
leogao
1y
14
23
Thoughts on gradient hacking
Richard_Ngo
1y
12
9
Alex Turner's Research, Comprehensive Information Gathering
adamShimi
1y
3
101
Gradient hacking
evhub
3y
39