用相似性学习改进前向-前向算法，提升准确率并减少推理次数
Reshaping the Forward-Forward Algorithm with a Similarity-Based Objective
- 引入基于相似性的损失函数，合并两次前向传播
- 在CIFAR-10上达56.22%准确率，接近反向传播的57.63%
- 适合追求生物合理性与高效推理的神经网络研究者
反向传播是支撑人工神经网络成功的关键算法，但存在生物不可行的反向锁死和全局误差传播等缺陷。为克服这些限制，前向-前向算法被提出，以额外的前向传播替代反向传播，更具生物合理性。然而该算法准确率显著低于反向传播，且最优形式因需多次前向传播而推理效率低。本文通过将其与相似性学习框架结合，重塑前向-前向算法，消除了推理阶段的多次前向传播需求。所提算法名为基于相似性三元组损失的前向-前向算法（FAUST）。在MNIST、Fashion-MNIST和CIFAR-10数据集上的实证评估表明，FAUST显著提升了准确率，缩小了与反向传播的差距。在CIFAR-10上，使用简单多层感知机架构即达56.22%准确率，接近反向传播基准57.63%。
原文摘要 · Abstract (English)
Backpropagation is the pivotal algorithm underpinning the success of artificial neural networks, yet it has critical limitations such as biologically implausible backward locking and global error propagation. To circumvent these constraints, the Forward-Forward algorithm was proposed as a more biologically plausible method that replaces the backward pass with an additional forward pass. Despite this advantage, the Forward-Forward algorithm significantly trails backpropagation in accuracy, and its optimal form exhibits low inference efficiency due to multiple forward passes required. In this work, the Forward-Forward algorithm is reshaped through its integration with similarity learning frameworks, eliminating the need for multiple forward passes during inference. This proposed algorithm is named Forward-Forward Algorithm Unified with Similarity-based Tuplet loss (FAUST). Empirical evaluations on MNIST, Fashion-MNIST, and CIFAR-10 datasets indicate that FAUST substantially improves accuracy, narrowing the gap with backpropagation. On CIFAR-10, FAUST achieves 56.22\% accuracy with a simple multi-layer perceptron architecture, approaching the backpropagation benchmark of 57.63\% accuracy.
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。