Search Results - "Malatesta, Enrico M."

The twin peaks of learning neural networks by Demyanenko, Elizaveta, Feinauer, Christoph, Malatesta, Enrico M, Saglietti, Luca

“…Recent works demonstrated the existence of a double-descent phenomenon for the generalization error of neural networks, where highly overparameterized models…”
Get full text

Properties of the Geometry of Solutions and Capacity of Multilayer Neural Networks with Rectified Linear Unit Activations by Baldassi, Carlo, Malatesta, Enrico M., Zecchina, Riccardo

“…Rectified linear units (ReLUs) have become the main model for the neural units in current deep learning systems. This choice was originally suggested as a way…”
Get full text

Unveiling the Structure of Wide Flat Minima in Neural Networks by Baldassi, Carlo, Lauditi, Clarissa, Malatesta, Enrico M, Perugini, Gabriele, Zecchina, Riccardo

“…The success of deep learning has revealed the application potential of neural networks across the sciences and opened up fundamental theoretical problems. In…”
Get full text

Star-Shaped Space of Solutions of the Spherical Negative Perceptron by Annesi, Brandon Livio, Lauditi, Clarissa, Lucibello, Carlo, Malatesta, Enrico M, Perugini, Gabriele, Pittorino, Fabrizio, Saglietti, Luca

“…Empirical studies on the landscape of neural networks have shown that low-energy configurations are often found in complex connected structures, where…”
Get full text

Two-loop corrections to large order behavior of φ4 theory by Malatesta, Enrico M., Parisi, Giorgio, Rizzo, Tommaso

“…We consider the large order behavior of the perturbative expansion of the scalar φ4 field theory in terms of a perturbative expansion around an instanton…”
Get full text

Two-loop corrections to large order behavior of φ 4 theory by Malatesta, Enrico M., Parisi, Giorgio, Rizzo, Tommaso

Get full text

High-dimensional manifold of solutions in neural networks: insights from statistical physics by Malatesta, Enrico M

“…In these pedagogic notes I review the statistical mechanics approach to neural networks, focusing on the paradigmatic example of the perceptron architecture…”
Get full text

Properties of the geometry of solutions and capacity of multi-layer neural networks with Rectified Linear Units activations by Baldassi, Carlo, Malatesta, Enrico M, Zecchina, Riccardo

“…Phys. Rev. Lett. 123, 170602 (2019) Rectified Linear Units (ReLU) have become the main model for the neural units in current deep learning systems. This choice…”
Get full text

Exact full-RSB SAT/UNSAT transition in infinitely wide two-layer neural networks by Annesi, Brandon L, Malatesta, Enrico M, Zamponi, Francesco

“…We analyze the problem of storing random pattern-label associations using two classes of continuous non-convex weights models, namely the perceptron with…”
Get full text

The twin peaks of learning neural networks by Demyanenko, Elizaveta, Feinauer, Christoph, Malatesta, Enrico M, Saglietti, Luca

“…Recent works demonstrated the existence of a double-descent phenomenon for the generalization error of neural networks, where highly overparameterized models…”
Get full text

Random Combinatorial Optimization Problems: Mean Field and Finite-Dimensional Results by Malatesta, Enrico M

“…This PhD thesis is organized as follows. In the first two chapters I will review some basic notions of statistical physics of disordered systems, such as…”
Get full text

Typical and atypical solutions in non-convex neural networks with discrete and continuous weights by Baldassi, Carlo, Malatesta, Enrico M, Perugini, Gabriele, Zecchina, Riccardo

“…We study the binary and continuous negative-margin perceptrons as simple non-convex neural network models learning random rules and associations. We analyze…”
Get full text

Impact of dendritic non-linearities on the computational capabilities of neurons by Lauditi, Clarissa, Malatesta, Enrico M, Pittorino, Fabrizio, Baldassi, Carlo, Brunel, Nicolas, Zecchina, Riccardo

“…Multiple neurophysiological experiments have shown that dendritic non-linearities can have a strong influence on synaptic input integration. In this work we…”
Get full text

Random Features Hopfield Networks generalize retrieval to previously unseen examples by Kalaj, Silvio, Lauditi, Clarissa, Perugini, Gabriele, Lucibello, Carlo, Malatesta, Enrico M, Negri, Matteo

“…It has been recently shown that a learning transition happens when a Hopfield Network stores examples generated as superpositions of random features, where new…”
Get full text

Instantons in $\phi^4$ Theories: Transseries, Virial Theorems and Numerical Aspects by Giorgini, Ludovico T, Jentschura, Ulrich D, Malatesta, Enrico M, Rizzo, Tommaso, Zinn-Justin, Jean

“…Phys.Rev.D 110 (2024) 036003 We discuss numerical aspects of instantons in two- and three-dimensional $\phi^4$ theories with an internal $O(N)$ symmetry group,…”
Get full text

The star-shaped space of solutions of the spherical negative perceptron by Annesi, Brandon Livio, Lauditi, Clarissa, Lucibello, Carlo, Malatesta, Enrico M, Perugini, Gabriele, Pittorino, Fabrizio, Saglietti, Luca

“…Empirical studies on the landscape of neural networks have shown that low-energy configurations are often found in complex connected structures, where…”
Get full text

Fluctuations in the random-link matching problem by Malatesta, Enrico M, Parisi, Giorgio, Sicuro, Gabriele

“…Phys. Rev. E 100, 032102 (2019) Using the replica approach and the cavity method, we study the fluctuations of the optimal cost in the random-link matching…”
Get full text

Wide flat minima and optimal generalization in classifying high-dimensional Gaussian mixtures by Baldassi, Carlo, Malatesta, Enrico M, Negri, Matteo, Zecchina, Riccardo

“…We analyze the connection between minimizers with good generalizing properties and high local entropy regions of a threshold-linear classifier in Gaussian…”
Get full text

Learning through atypical "phase transitions" in overparameterized neural networks by Baldassi, Carlo, Lauditi, Clarissa, Malatesta, Enrico M, Pacelli, Rosalba, Perugini, Gabriele, Zecchina, Riccardo

“…Current deep neural networks are highly overparameterized (up to billions of connection weights) and nonlinear. Yet they can fit data almost perfectly through…”
Get full text

Unveiling the structure of wide flat minima in neural networks by Baldassi, Carlo, Lauditi, Clarissa, Malatesta, Enrico M, Perugini, Gabriele, Zecchina, Riccardo

“…The success of deep learning has revealed the application potential of neural networks across the sciences and opened up fundamental theoretical problems. In…”
Get full text