Sparse In-Network Learning via Shortest-Path Backpropagation and Finite-Rate Gating

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Salehi, Mohammad Reza Deylam
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913155424190464
author Salehi, Mohammad Reza Deylam
author_facet Salehi, Mohammad Reza Deylam
contents In-network learning (INL) trains distributed neural modules by exchanging latent activations and backpropagated errors over a communication graph. This letter proposes Dijkstra-pruned INL (D-INL), which removes non-tree links by retaining a capacity-aware shortest-path tree rooted at the fusion node. To balance sparsity and predictive information, local routing (or aggregation) is modeled as a finite-rate stochastic gate with rate $R_g=I(Z; T)$. We derive a rate-distortion-generalization bound and validate the method on a reproducible distributed-classification experiment, where D-INL reduces training exchange by $70.4\%$ while preserving accuracy within the standard deviation of dense INL. Adding finite-rate regularization further reduces the estimated latent rate by $45.7\%$ relative to unregularized Dijkstra INL.
format Preprint
id arxiv_https___arxiv_org_abs_2605_23424
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Sparse In-Network Learning via Shortest-Path Backpropagation and Finite-Rate Gating
Salehi, Mohammad Reza Deylam
Information Theory
Machine Learning
In-network learning (INL) trains distributed neural modules by exchanging latent activations and backpropagated errors over a communication graph. This letter proposes Dijkstra-pruned INL (D-INL), which removes non-tree links by retaining a capacity-aware shortest-path tree rooted at the fusion node. To balance sparsity and predictive information, local routing (or aggregation) is modeled as a finite-rate stochastic gate with rate $R_g=I(Z; T)$. We derive a rate-distortion-generalization bound and validate the method on a reproducible distributed-classification experiment, where D-INL reduces training exchange by $70.4\%$ while preserving accuracy within the standard deviation of dense INL. Adding finite-rate regularization further reduces the estimated latent rate by $45.7\%$ relative to unregularized Dijkstra INL.
title Sparse In-Network Learning via Shortest-Path Backpropagation and Finite-Rate Gating
topic Information Theory
Machine Learning
url https://arxiv.org/abs/2605.23424