arXiv：cs.LG· Mustafa Uzun, Mete Erdogan, Cengiz Pehlevan, Alper T. Erdogan·· 4 小时前AI 评分36
Score Broadcast and Decorrelation：面向通用可微损失的广播式信用分配框架
Score Broadcast and Decorrelation: A General Framework for Broadcast-Based Credit Assignment
AI 导读
研究者提出 Score Broadcast and Decorrelation（SBD）框架，将广播式信用分配推广到交叉熵、Bregman 散度、proper scoring rules 和指数族负对数似然等通用可微损失。
正文
Abstract:We introduce Score Broadcast and Decorrelation (SBD), a principled framework for broadcast-based credit assignment for general families of differentiable losses. Error broadcast is a biologically plausible alternative to backpropagation that sends output information to hidden layers without weight transport. The Error Broadcast and Decorrelation (EBD) framework, recently introduced for the mean-squared-error (MSE) setting, grounded this mechanism in the stochastic orthogonality of optimal estimators, under which the optimal residual is orthogonal to functions of the input. We generalize that foundation by introducing an orthogonality principle between the output score (the gradient of loss with respect to the final-layer output) and hidden-layer activations, which holds whenever the optimal score has conditional mean zero. This single principle unifies broadcast-based credit assignment across the standard differentiable-loss families, including cross-entropy, Bregman divergences, proper scoring rules, and exponential-family negative log-likelihoods. The framework supplies a theoretical grounding for the three-factor learning rule under general losses, with the neuromodulatory factor derived as the broadcast loss score. We derive the cross-entropy case explicitly, characterize the admissible loss class, and introduce a score vector expansion technique that enriches the broadcast signal while preserving the orthogonality framework. Experiments on CIFAR-10 and Tiny ImageNet show that SBD substantially improves over existing broadcast approaches, with score vector expansion delivering further gains. Overall, this work identifies the loss score as the signal to broadcast, supplies the orthogonality theory and theoretical grounding for the three-factor learning rule from neuroscience, and shows how score vector expansion enriches the decorrelation directions of the resulting objective.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.30638 [cs.LG] |
| (or arXiv:2605.30638v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.30638 arXiv-issued DOI via DataCite |
Submission history
From: Mustafa Uzun [view email]
[v1]
Thu, 28 May 2026 22:38:28 UTC (3,650 KB)
[v2]
Wed, 7 Oct 2026 13:42:26 UTC (4,712 KB)
来源：arXiv：cs.LG · arxiv.org