Yongfei Guo, Xuan Sun, Ang Li, Zhao Zhang, Yuanhao Gong · 2026 · arXiv
DOI: 10.48550/arXiv.2610.29697 · open-access PDF
La retropropagación sigue siendo el paradigma de entrenamiento dominante, utilizando gradientes retroactivos y actualizaciones locales que pueden sufrir de la desaparición o explosión de gradientes en redes profundas. Aunque el algoritmo Avanzado-Avanzado evita la retropropagación convencional, sigue siendo principalmente local o de capa, motivando el aprendizaje global de la red entera utilizando sólo evaluaciones avanzadas. Para las redes neuronales de wavelet de núcleo compartido, por lo tanto, proponemos un marco global de aprendizaje solo hacia adelante basado en CMA-ES.
Machine translation (Meta M2M100); the English original follows.
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 searc
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.29697, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange