2 papers
cond-mat.dis-nn2025
Overparametrization bends the landscape: BBP transitions at initialization in simple Neural Networks
Brandon Livio Annesi, Dario Bocchi, Chiara Cammarota
High-dimensional non-convex loss landscapes play a central role in the theory of Machine Learning. Gaining insight into how these landscapes interact with gradient-based optimizati…
cond-mat.dis-nn2025
Exact full-RSB SAT/UNSAT transition in infinitely wide two-layer neural networks
Brandon L. Annesi, Enrico M. Malatesta, Francesco Zamponi
We analyze the problem of storing random pattern-label associations using two classes of continuous non-convex weights models, namely the perceptron with negative margin and an inf…