Evolutionary Connectionism

Proposed framework translating connectionist learning principles into natural selection domain to explain ETIs.

Neighborhood — ranked by edge-count

paper

thinker

Michael Levin
implements
Richard Watson
introduces
Co-author; Electronics and Computer Science/Institute for Life Sciences, University of Southampton; develops connectionist frameworks for collective intelligence.
David Power
associated_with
Dniel Czgel
associated_with

method

Unsupervised Learning
implements
Learning that builds a low-dimensional model of input data without error signals or rewards; Hebbian learning is an example.

concept

Deep Learning
associated_withimplements
Learning hierarchical representations of non-decomposable functions; proposed as formal equivalent to ETI process.
Evolutionary Transitions In Individuality
implements
Non-aggregative fitness interactions
cites

community

claim

Evolution is more intelligent than we realised.
supports
Watson's reinterpretation of formal equivalence between evolution and learning, beyond random variation framework.
Organised structure of relationships between parts, sufficient to exhibit information integration and collective action, can be produced via fully-distributed unsupervised learning.
supports
Central claim from connectionist models: complex coordination emerges without centralized control or external teacher.

framework

Connectionist Models
extends
Neural network models demonstrating how organized functional relationships emerge via unsupervised learning; basis for evolutionary connectionism analogy.
Evolutionary Transitions in Individuality (ETI)
implements
Theory explaining how new levels of biological organization and individuality emerge through transitions in collective intelligence and problem-solving rescaling.
Connectionist Models of Cognition and Learning
extends
Models where intelligence arises from organisation of connections between simple processing units, used as basis for evolutionary connectionism

question

How can selection explain adaptations at system level before the higher-level unit exists as an evolutionary unit?
answered_by
Fundamental challenge: conventional evolutionary theory cannot explain system-level features required for ETI without presupposing the higher-level unit.

hypothesis

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.