Reparameterizing layer weights in a learned orthonormal eigenbasis is claimed to improve ImageNet classification, cross-modal retrieval, and enable a faster backpropagation-free variant that surpasses standard backpropagation.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Eigen Neural Network: Unlocking Generalizable Vision with Eigenbasis
Reparameterizing layer weights in a learned orthonormal eigenbasis is claimed to improve ImageNet classification, cross-modal retrieval, and enable a faster backpropagation-free variant that surpasses standard backpropagation.