LESSWRONG
LW

701
Wikitags

Emergent Behavior ( Emergence )

Edited by markov, et al. last updated 26th Oct 2022

Emergent Behavior or emergence occurs when an entity is observed to have certain properties , that its individual parts do not have on their own. Certain properties or behaviors might come about only when the parts that make up that entity interact as a wider whole , or when the number of parts making up that entity reach a certain minimum threshold.

It is hypothesized that an AI system might display emergent properties simply as a function of scale. This is supported by the fact that various complex systems exist in nature that have displayed such emergent properties due to quantitative differences. Examples include :

  • Heart : Individually cells cannot pump bood whereas the whole heart can.
  • Uranium : Small amounts, nothing special, large amounts = nuclear reactions
  • Civilization : Individuals, nothing special, collective specialization = human civilization

This raises concerns that due to the quantitative increases in the size of AI models , we might soon hit an unexpected threshold and see unexpected differences in qualitiative behaviors and capabilites. This provides motivations to study complex systems with emergent phenomena to be able to draw inferences for when , and what types of capabilites might emerge.

References 

  • More is Different for AI Sequence
  • Wikipedia
Subscribe
Discussion
1
Subscribe
Discussion
1
Posts tagged Emergent Behavior ( Emergence )
128The Futility of Emergence
Eliezer Yudkowsky
18y
142
6Crisis of Faith case study: beyond reductionism?
MalcolmOcean
2y
9
164A Bird's Eye View of the ML Field [Pragmatic AI Safety #2]
Ω
Dan H, TW123
3y
Ω
8
80Introduction to Pragmatic AI Safety [Pragmatic AI Safety #1]
Ω
Dan H, TW123
3y
Ω
3
10A Case for Conscious Significance rather than Free Will.
James Stephen Brown
1y
2
53Are Emergent Abilities of Large Language Models a Mirage? [linkpost]
Matthew Barnett
2y
21
34Unexpected Conscious Entities
Gunnar_Zarncke
5mo
7
25Emergent Abilities of Large Language Models [Linkpost]
aog
3y
2
154Matt Botvinick on the spontaneous emergence of learning algorithms
Ω
Adam Scholl
5y
Ω
87
113Complex Behavior from Simple (Sub)Agents
moridinamael
6y
14
76Selling Nonapples
Eliezer Yudkowsky
17y
78
51Perform Tractable Research While Avoiding Capabilities Externalities [Pragmatic AI Safety #4]
Ω
Dan H, TW123
3y
Ω
3
28Emergence Spirals—what Yudkowsky gets wrong
James Stephen Brown
4mo
25
27Bing finding ways to bypass Microsoft's filters without being asked. Is it reproducible?
Christopher King
3y
15
24Interpretability through two lenses: biology and physics
raphael
2mo
4
Load More (15/35)
Add Posts