Limits RegistryPUBLIC RECORDLimits RegistryResearch Console →Arbitrarily deep residual networks with one-neuron hidden layers are universal L1 approximators.
The question, scope, and sources behind this Registry record.
Under the recorded neural network theory model, establish one hidden neuron per residual block suffices.
Current frontiers derived from accepted Claims.
The accepted equality closes this optimization result.
Assertions tied to evidence, attribution, and review.
The frontier as it changed over time.
Only accepted Claims matching the current specification contribute to the displayed bounds. Strict inequalities remain open; contradictory Claims require editorial review.
No accepted machine-checked reproductions are recorded for this Limit.
Limits Registry. LR-003617. One-neuron residual-block universality. 2026.