claim

active

claim:bottom-up-learning-creates-a-non-decomposable-whole-attractors-that-are-non-linearly-separable-functions-of-the-inputs-and-depend-on-the-system-s-own-internal-history-which-means-that-credit-assignment-or-reward-at-the-level-of-individual-parts-becomes-ineffective

Bottom-up learning creates a non-decomposable whole (attractors that are non-linearly separable functions of the inputs and depend on the system’s own internal history), which means that credit assignment or reward at the level of individual parts becomes ineffective.

Explains how collective cognition becomes irreducible to parts.

Source paper

extracted_from

The collective intelligence of evolution and development

(2023) · Watson, Richard · Levin, Michael

Neighborhood — ranked by edge-count

Claims (1)

claim

Reinforcement learning acting on individual characteristics affecting their connections to others can result in dynamics that are equivalent to unsupervised learning at the system scale.
extends
Key insight linking individual rewards to system-level learning.

Related by similarity (8)

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.

The conditions necessary to evolve a new level of individuality are described by the conditions necessary to learn non-decomposable functions (deep model induction)claim0.791
Core interpretive claim of the paper connecting ETIs to connectionist learning
Connectionist models of cognition and learning identify conditions where collective intelligence can arise bottom-up, using only distributed learning mechanisms without system-level or global feedback.claim0.786
Central claim about the power of connectionism.
Connectionist models can exhibit learning bottom-up, without centralised control or an external teacher, and without any performance feedback applied at the system level.claim0.782
Key property of distributed unsupervised learning.
Conditions necessary to evolve new individuality level are described by conditions for learning non-decomposable functions (deep model induction)hypothesis0.775
Associative learning occurs in gene regulatory networks and other non-neural systems, making it substrate-independent.claim0.773
Learning neural networks can enable 'chunking' and rescale problem-solving to higher organizational levels, a mechanism intrinsic to transitions in individuality.hypothesis0.770
Bottom-up incremental adjustment of relationships can increase system-level welfare in a manner functionally equivalent to distributed learning.claim0.766
How short-sighted agents improve global outcomes through connection modification.
Unsupervised learning builds a low-dimensional model of the input data.claim0.759
Clarifies what unsupervised learning does.