Tag
This paper introduces spectral updates for local learning that enhance depth robustness and reduce hyperparameter sensitivity, achieving better accuracy than local Adam on CIFAR-10 benchmarks.
A side project presents a Hebbian architecture AI model that avoids backpropagation and gradients, achieving 50 epochs on CIFAR-10 with emergent behaviors like accuracy dips followed by jumps and recovery after targeted damage.
This paper introduces GOEN, a pipeline combining multi-scale features, L2 normalization, and Mahalanobis distance for OOD detection, and finds that CenterLoss regularization actually degrades OOD performance despite improving classification accuracy.