Back to search

Article

Native Nonlinearities in Equilibrium-Propagation-Trained Wave-Field Classifiers: A Linear-Subspace Diagnostic

2026-06-03

Abstract excerpt

<title>Abstract</title> <p>Local learning rules tune a physical substrate to improve task performance, but they also implicitly select an operating regime—the subset of equations of motion used at inference. For wave-field classifiers trained by equilibrium propagation (EP), this regime is uncharacterized. We probe it for a three-channel Landau–Ginzburg classifier trained on TIMIT 40-class phoneme recognition. Th...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work
857907aa-db73-5ee6-96c2-b4ffe3dd4fdb
DOI
10.21203/rs.3.rs-9897534/v1
Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest
Interactive article relationship graphSelect a related publication card to move it into the centre and load its closest explainable connections. Solid lines are source-backed structured connections. Dashed lines are semantic discovery signals and are not scientific evidence.
Native Nonlinearities in Equilibrium-Propagation-Trained Wave-Field Classifiers: A Linear-Subspace DiagnosticDOI 10.21203/rs.3.rs-9897534/v1
Select a neighboring publication to make it the new centre.