For ReLU networks with width at least two in input and hidden layers, an open set of parameters is identifiable, implying functional dimension equals parameter count minus hidden neurons.
Daniel Kunin, Javier Sagastuy-Brena, Surya Ganguli, Daniel LK Yamins, and Hidenori Tanaka
3 Pith papers cite this work, alongside 6 external citations. Polarity classification is still indexing.
3
Pith papers citing it
6
external citations · OpenAlex
fields
cs.LG 3years
2026 3verdicts
UNVERDICTED 3representative citing papers
Complete characterization of generic parameter fibers and symmetries in three-layer bottleneck ReLU networks with a polynomial-time functional equivalence test.
Shallow ReLU networks admit a complete classification of parameter symmetries obtained by exploiting ReLU non-differentiability rather than analytic activation assumptions.
citing papers explorer
-
Most ReLU Networks Admit Identifiable Parameters
For ReLU networks with width at least two in input and hidden layers, an open set of parameters is identifiable, implying functional dimension equals parameter count minus hidden neurons.
-
The Symmetries of Three-Layer ReLU Networks
Complete characterization of generic parameter fibers and symmetries in three-layer bottleneck ReLU networks with a polynomial-time functional equivalence test.
-
A Complete Symmetry Classification of Shallow ReLU Networks
Shallow ReLU networks admit a complete classification of parameter symmetries obtained by exploiting ReLU non-differentiability rather than analytic activation assumptions.