Hopfield Network

A recurrent connectionist architecture that implements associative memory and distributed computation through symmetric weighted connections and Hebbian learning rules. The network converges to stable states through recurrent dynamics, enabling both memory retrieval and combinatorial problem-solving in a fully distributed manner.

Neighborhood — ranked by edge-count

Thinkers (2)

thinker

John Hopfield
associated_withintroduces
Alexis Pietak
studies

Methods (1)

method

Hebbian Learning
implements
Principle that correlations strengthen connections; implements distributed learning in connectionist networks without centralized supervision.

Concepts (3)

concept

Connectionism
implements
Central framework proposing intelligence resides in organization of relationships between components, not in individual parts; used to unify individual and collective intelligence.
Self-attention
analogous_to
A form of key-query attention within a single input sequence; core to Transformers.
Hopfield Networks is All You Need (Ramsauer et al., 2020)
cites
Paper showing Hopfield networks are closely related to transformers; key intermediary result used to connect TEM to transformers.

Frameworks (2)

framework

Connectionist Models
implements
Neural network models demonstrating how organized functional relationships emerge via unsupervised learning; basis for evolutionary connectionism analogy.
Tolman-Eichenbaum Machine (TEM)
implements
Neuroscience model of hippocampal formation that the paper shows is mathematically equivalent to a transformer with recurrent position encodings.

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.

Hopfield network can store multiple patterns, recall them via content-addressable memory, and generalise to novel patterns with the same underlying structure.finding0.744
Summary of known Hopfield network capabilities used as a model for collective computation.
Hopfield Network Memory Capacity Scalingconcept0.737
Technical issue: original Hopfield networks scale linearly with neuron count; exponential activations enable 2^(N/2) scaling but softmax used in TEM-t has intermediate properties.
Highway Networksconcept0.712
Early Schmidhuber group architecture using residual connections, precursor to the residual stream structure in transformers
Feed-forward networksconcept0.688
Networks without feedback loops, such as the Perceptron; limited to linearly separable functions.
Flow networksconcept0.680
Networks for transport via fluid flow (biological vascular, engineered microfluidic); pipe properties adapt locally to control flow conductance and enable learning
Bioelectric Networksframework0.679
Ancient mechanism using ion channels, gap junctions, and neurotransmitters as 'cognitive glue' enabling collective intelligence across morphogenesis and behavior.
Elastic networksconcept0.674
Spring networks and mechanical systems used to demonstrate physical learning through modifications of bond stiffnesses in response to strain
Reversible Residual Network (RevNet)method0.674
Bijective invertible architecture used to implement non-linear alignment maps ϕ_nonlin