提出误差广播与去相关机制，让神经网络更高效地学习
Error Broadcast and Decorrelation as a Potential Artificial and Natural Learning Mechanism
- 用输出误差直接广播到各层，避免反向传播的权重传输
- 在基准数据集上表现优于或媲美现有误差广播方法
- 机制符合生物可解释性，适合研究类脑学习的人看
我们提出一种名为误差广播与去相关（Error Broadcast and Decorrelation, EBD）的新颖神经网络学习框架，通过将输出误差直接广播至各层来解决信用分配问题，从而规避反向传播中的权重传输。EBD 基于最小均方误差估计器的随机正交性原理——最优估计器的误差与输入函数正交。基于此，EBD 定义了逐层损失函数，直接惩罚层激活与输出误差之间的相关性，为误差广播提供了理论基础。这一机制自然导出实验观察到的三因素学习规则，并可融入生物合理框架以提升性能与合理性。数值实验表明，EBD 在基准数据集上的表现与其它误差广播方法相当或更优。研究结果确立了 EBD 作为高效、生物合理且理论严谨的神经网络训练替代方案。实现代码已公开：https://github.com/meterdogan07/error-broadcast-decorrelation。
原文摘要 · Abstract (English)
We introduce Error Broadcast and Decorrelation (EBD), a novel learning framework for neural networks that addresses credit assignment by directly broadcasting output errors to individual layers, circumventing weight transport of backpropagation. EBD is rigorously grounded in the stochastic orthogonality property of Minimum Mean Square Error estimators. This fundamental principle states that the error of an optimal estimator is orthogonal to functions of the input. Guided by this insight, EBD defines layerwise loss functions that directly penalize correlations between layer activations and output errors, thereby establishing a principled foundation for error broadcasting. This theoretically sound mechanism naturally leads to the experimentally observed three-factor learning rule and integrates with biologically plausible frameworks to enhance performance and plausibility. Numerical experiments demonstrate EBD's competitive or better performance against other error-broadcast methods on benchmark datasets. Our findings establish EBD as an efficient, biologically plausible, and principled alternative for neural network training. The implementation is available at: https://github.com/meterdogan07/error-broadcast-decorrelation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。