Target Factorization: Loss Geometry and Component Learning Dynamics

A class label can combine attributes, such as shape and color. We ask whether predicting these attributes separately makes each easier to learn. For cross-entropy, additive component scores remove an exact cross-component curvature term from the loss. This geometric separation does not guarantee faster learning. In a binary model, learned interactions slow additive-coordinate growth while accelerating predictive confidence: locally at unequal positive rates and at every confidence level above chance at unit rates. In a matched Boolean experiment with 12 initializations, a component reaches an accuracy threshold later when paired with an easier companion target. Allowing interactions reduces this delay from 24.00 to 14.58 updates (paired difference −9.42; exploratory 95% bootstrap interval [−14.33, −4.33]). Integrated prediction loss and error show the same direction without an accuracy threshold. Predictive dependence and backbone gradient alignment do not establish a causal explanation. A parity-target grid retains positive joint-versus-isolated training contrasts under exact output factorization, although its heads are not fully matched. Separating predictions therefore need not make their learning independent. Comparisons of output encodings should measure component-level predictions under explicit training controls.

Authors

Publication Details

Journal
Preprints.org
Published
2026-11-16
DOI
https://doi.org/10.20944/preprints202611.1529.v1
Primary Topic
Neural Networks and Applications
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Target Factorization: Loss Geometry and Component Learning Dynamics

Yunguo Yu
1 links
Preprints.org
Neural Networks and Applications
preprint

Target Factorization: Loss Geometry and Component Learning Dynamics

Yunguo Yu
preprint en

Abstract

A class label can combine attributes, such as shape and color. We ask whether predicting these attributes separately makes each easier to learn. For cross-entropy, additive component scores remove an exact cross-component curvature term from the loss. This geometric separation does not guarantee faster learning. In a binary model, learned interactions slow additive-coordinate growth while accelerating predictive confidence: locally at unequal positive rates and at every confidence level above chance at unit rates. In a matched Boolean experiment with 12 initializations, a component reaches an accuracy threshold later when paired with an easier companion target. Allowing interactions reduces this delay from 24.00 to 14.58 updates (paired difference −9.42; exploratory 95% bootstrap interval [−14.33, −4.33]). Integrated prediction loss and error show the same direction without an accuracy threshold. Predictive dependence and backbone gradient alignment do not establish a causal explanation. A parity-target grid retains positive joint-versus-isolated training contrasts under exact output factorization, although its heads are not fully matched. Separating predictions therefore need not make their learning independent. Comparisons of output encodings should measure component-level predictions under explicit training controls.

Preprints.org
Neural Networks and Applications
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon