3 papers
cond-mat.dis-nn2025
Dynamical Learning in Deep Asymmetric Recurrent Neural Networks
Davide Badalotti, Carlo Baldassi, Marc Mézard +2
We investigate recurrent neural networks with asymmetric interactions and demonstrate that the inclusion of self-couplings or sparse excitatory inter-module connections leads to th…
cs.LG2025
Learning by Steering the Neural Dynamics: A Statistical Mechanics Perspective
Mattia Scardecchia
Despite the striking successes of deep neural networks trained with gradient-based optimization, these methods differ fundamentally from their biological counterparts. This gap rai…
cs.CV2025
Unsupervised Transformer Pre-Training for Images: Self-Distillation, Mean Teachers, and Random Crops
Mattia Scardecchia
Recent advances in self-supervised learning (SSL) have made it possible to learn general-purpose visual features that capture both the high-level semantics and the fine-grained spa…