concept
active
concept:on-a-model-of-associative-memory-with-huge-storage-capacity-demircigil-et-al-2017On a Model of Associative Memory with Huge Storage Capacity (Demircigil et al., 2017)
Analytical result showing exponential power activation allows memory storage scaling as 2^(N/2); cited in context of Hopfield scaling.
Neighborhood — ranked by edge-count
Papers (1)
paper
Concepts (1)
concept
- 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.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Large Associative Memory Problem in Neurobiology and Machine Learning (Krotov & Hopfield, 2020)concept0.832Paper proposing neurally plausible instantiation of Hopfield/transformer with feature neurons and memory neurons; used to derive place cell interpretation.
- Paper exploring two-type feature binding in dense associative memory; analogous to TEM-t binding g and x in memory neurons.
- Piumarta argues the primary practical challenge is storage management: when any key becomes unreachable, all values indexed by that key must be deleted, requiring cooperation between primitive mechanism and garbage collector.
- Memory system that stores patterns in connection weights and recalls them from partial or noisy cues; property of Hopfield networks and evolved networks
- The capacity for unlimited associative learning is not a good indicator for consciousness in AI.claim0.764Rejecting UAL as a reliable indicator for artificial systems.