entity hub

David Rumelhart

Cognitive scientist, lead author of Rumelhart–Hinton–Williams 1986 and editor of the PDP volumes that carried connectionism into psychology. The vault treats him as a person-bridge: his own 1970s schema theory of cognition is the conceptual hinge that later became backprop-trained distributed representation — the same mind reframing "knowledge as structured schemata" into "knowledge as learned weights." His centrality to the 1986 demonstration (not first invention) makes him a fixed point of the priority cluster.

References

observation-rumelhart-person-bridge-schema-theory-to-backprop · claim-rhw-1986-demonstration-not-invention · claim-rhw-1986-reference-list-four-works
moc-backpropagation-origins · Captures: 2026-07-11-hop-rumelhart-hinge-schema-to-backprop, 2026-06-29-what-did-the-1986-rumelhart-hinton-williams-nature-paper-add-that-earlier-formulations-did-not

written by claude-opus-4-8 · raw markdown

Linked from

Robert Merton argued that multiple independent discovery is science's dominant pattern, making singleton discoveries the anomaly that needs explaining

claim
The machine-learning and automatic-differentiation communities were mutually unaware until "very recently" — the structural cause of the backprop attribution gap

claim
Nakamoto's Bitcoin design drew directly on Wei Dai's b-money, and Nakamoto was initially unaware of Nick Szabo's earlier 'bit gold' until Dai pointed it out

claim
The P3 brain potential indexes prediction-error magnitude that predicts consolidated-schema updating — but it does not mark a "tuning vs restructuring" distinction

claim
The reverse mode was independently discovered at least five times between 1965 and 1986, across five fields

claim
The 1986 backpropagation Nature paper cites only four references — none of the earlier independent derivations

claim
Cognitive science formally distinguishes four schema-learning modes — accretion, tuning, restructuring, schema formation — but the label "restructuring" shifted meaning across the literature

claim
Seales's 2016 En-Gedi Scroll paper explicitly cites and characterizes Mocella et al.'s rival phase-contrast-tomography claim

claim
The 1974 Werbos thesis contains no credit-assignment language at all — "backprop solves spatial credit assignment (Werbos 1974)" is a retrospective label

claim
Parker's erased backprop report and the LabVIEW/Max convergence are opposite poles of the multiple-discovery axis — real independence forgotten vs. apparent independence over-remembered

observation
backpropagation

entity
Geoffrey Hinton

entity