Backpropagation began as an attempt to turn Freud’s “psychic energy” into math
fkasummer · x · 2026-07-22
Backprop started as an attempt to mathematize Freud’s “psychic energy”
The quoted story says backpropagation did not begin purely as a supervised-learning trick. It emerged from an attempt to translate Freud’s idea of “psychic energy” into math.
- The speaker recalls writing a prospectus on an intelligence model with an adaptive critic.
- Backprop was part of that system, but the goal was to model a flow of psychic energy through the network.
- He says he converted the idea into derivative equations and an adaptive critic that backpropagated to a critic.
- The anecdote is used to explain how a concept now central to modern AI had a surprisingly unusual intellectual origin.
More from Research
- RAND publishes first roadmap for protecting valuable algorithmic know-how — Scobleizer · 2026-07-22
- Shengshu says world models, not LLMs alone, will power video, robots and agents — 生数科技 · 2026-07-22
- SWE-Pruner Pro Shows Coding Agents Already Know What Context to Drop — rohanpaul_ai · 2026-07-22
- SkewAdam cuts MoE optimizer memory by 97.4% and fits 6.78B on a 40GB GPU — Kooky-Ad-4124 · 2026-07-22
- New paper defines four conditions that turn an LLM into a coding agent — alex_verem · 2026-07-22
- Spectral clustering method groups Markov chains via P² eigenvectors and weighted k-means — michaelchchoi · 2026-07-22