Newest papers

Forward-Only Learning for Shared-Kernel Wavelet Neural Networks without Backpropagation

Yongfei Guo, Xuan Sun, Ang Li, Zhao Zhang, Yuanhao Gong · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.29697 · cited by 0 works (OpenAlex) · open-access PDF

Portuguese

Embora o algoritmo Avançado evite o paradigma de treinamento dominante, usando gradientes atrasados e atualizações locais que podem sofrer de gradientes desaparecidos ou explodindo em redes profundas. Embora o algoritmo Avançado evite o backpropagamento convencional, permanece principalmente local ou de camada, motivando o aprendizado global de rede inteira usando apenas avaliações futuras. Para redes neurais de wavelet de núcleo compartilhado, propomos, portanto, um quadro de aprendizagem global baseado em CMA-ES. Nós introduzimos ainda a decomposição de parâmetros conscientes da estrutura e a atribuição de crédito específica do canal, factorizando a pesquisa em três subespaços. Em BSDS500, nosso método reduz o teste de MSE de 4.1783 para 4.0130 e melhora o PSNR de 42.

Machine translation (Meta M2M100); the English original follows.

Abstract

Backpropagation remains the dominant training paradigm, using backward gradients and local updates that can suffer from vanishing or exploding gradients in deep networks. Although the Forward-Forward algorithm avoids conventional backpropagation, it remains mainly local or layer-wise, motivating global whole-network learning using forward evaluations only. For shared-kernel wavelet neural networks, we therefore propose a CMA-ES-based global forward-only learning framework. We further introduce structure-aware parameter decomposition and channel-specific credit assignment, factorizing the search into three subspaces. On BSDS500, our method reduces test MSE from 4.1783 to 4.0130 and improves PSNR from 42.621 to 42.787~dB over Adam/BP without backward propagation. Beyond these gains, training requires only 6 training-set coverages, compared with 1500 epochs for BP. Ablations show that structural decomposition and aligned credit assignment are more important than increasing population size, demonstrating the feasibility of whole-network forward-only learning for compact reconstruction networks. These results suggest a potential route toward forward-only optimization of compact reconstruction models for resource-constrained imaging and edge or near-sensor systems.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Backpropagation, Artificial intelligence, Artificial neural network, Computer science, Wavelet, Pattern recognition (psychology)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange