arXiv subjects

Dario Bocchi

Publications and source records attributed to Dario Bocchi.

8 recordsLinked to original sources

arXiv2026

Spectral phase transitions and trainability in neural network learning dynamics

The emergence of low-dimensional structures in the spectra of neural network weight matrices is a common empirical feature of trained models, but the dynamical origin of this phenomenon during learning remains an open problem. We formulate neural network training as the stochastic evolution of an initially random matrix ensemble, driven by stochastic gradient descent (SGD) updates that reshape the spectral bulk while amplifying signal strength. This induces a Baik-Ben Arous-P\'ech\'e (BBP) transition during training, where isolated eigenvalues detach from the random bulk distribution, providing a dynamical framework for representation formation in high-dimensional learning dynamics. We demonstrate this in a solvable linear teacher-student model, where spectral evolution is analytically tractable and a phase diagram of trainability governed by the step size (or learning rate) and initial weight variance is obtained, and subsequently extend our formalism beyond the linear regime to nonlinear and stochastic settings. Numerical simulations in realistic settings support this picture, showing robust emergence of spectral alignment during training. Our results suggest that spectral analysis may provide a unified perspective of stochastic learning dynamics, linking trainability, optimisation hyperparameters, spectral phase transitions, and representation learning in neural networks.

cond-mat.dis-nn↗

arXiv2026

Discontinuous BBP transitions

The Baik-Ben Arous-Peche (BBP) transition sets fundamental limits for detecting low-rank structure in noisy high-dimensional data and underlies a wide range of spectral methods in many fields from physics to statistics and data sciences. In standard settings, this transition is continuous, implying that signal recovery emerges gradually above a sharp threshold. We show that BBP transitions can instead be discontinuous in very general settings and provide a full theory of this phenomenon. When the eigenvalue density vanishes faster than linearly at the spectral edge, the overlap between the leading eigenvector and the signal jumps discontinuously at the critical point. We study this mechanism in deformed Gaussian and reweighted Wishart ensembles. We analyze in detail the finite-size effects, which play a central and qualitatively new role in the discontinuous BBP transition. Unlike the continuous BBP transition, we establish the existence of an extended pre-critical region where informative eigenvectors emerge well before the asymptotic threshold. The main consequence-and difference from the continuous BBP transition-is that signal recovery can occur at significantly lower signal-to-noise ratio and it is accompanied by strong sample-to-sample variability. Our results show the relevance and the novelty of the discontinuous BBP transition, and highlight the practical implications for signal detection.

cond-mat.dis-nn↗

arXiv2026

Escape dynamics and implicit bias of one-pass SGD in overparameterized quadratic networks

We analyze the one-pass stochastic gradient descent dynamics of a two-layer neural network with quadratic activations in a teacher--student framework. In the high-dimensional regime, where the input dimension $N$ and the number of samples $M$ diverge at fixed ratio $\alpha = M/N$, and for finite hidden widths $(p,p^*)$ of the student and teacher, respectively, we study the low-dimensional ordinary differential equations that govern the evolution of the student--teacher and student--student overlap matrices. We show that overparameterization ($p>p^*$) only modestly accelerates escape from a plateau of poor generalization by modifying the prefactor of the exponential decay of the loss. We then examine how unconstrained weight norms introduce a continuous rotational symmetry that results in a nontrivial manifold of zero-loss solutions for $p>1$. From this manifold the dynamics consistently selects the closest solution to the random initialization, as enforced by a conserved quantity in the ODEs governing the evolution of the overlaps. Finally, a Hessian analysis of the population-loss landscape confirms that the plateau and the solution manifold correspond to saddles with at least one negative eigenvalue and to marginal minima in the population-loss geometry, respectively.

cond-mat.dis-nn↗

arXiv2025

Overparametrization bends the landscape: BBP transitions at initialization in simple Neural Networks

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 optimization methods, even in relatively simple models, can shed light on this enigmatic feature of neural networks. In this work, we will focus on a prototypical simple learning problem, which generalizes the Phase Retrieval inference problem by allowing the exploration of overparametrized settings. Using techniques from field theory, we analyze the spectrum of the Hessian at initialization and identify a Baik-Ben Arous-Péché (BBP) transition in the amount of data that separates regimes where the initialization is informative or uninformative about a planted signal of a teacher-student setup. Crucially, we demonstrate how overparameterization can bend the loss landscape, shifting the transition point, even reaching the information-theoretic weak-recovery threshold in the large overparameterization limit, while also altering its qualitative nature. We distinguish between continuous and discontinuous BBP transitions and support our analytical predictions with simulations, examining how they compare to the finite-N behavior. In the case of discontinuous BBP transitions strong finite-N corrections allow the retrieval of information at a signal-to-noise ratio (SNR) smaller than the predicted BBP transition. In these cases we provide estimates for a new lower SNR threshold that marks the point at which initialization becomes entirely uninformative.

cond-mat.dis-nn↗

arXiv2025

Pseudo-likelihood produces associative memories able to generalize, even for asymmetric couplings

Energy-based probabilistic models learned by maximizing the likelihood of the data are limited by the intractability of the partition function. A widely used workaround is to maximize the pseudo-likelihood, which replaces the global normalization with tractable local normalizations. Here we show that, in the zero-temperature limit, a network trained to maximize pseudo-likelihood naturally implements an associative memory: if the training set is small, patterns become fixed-point attractors whose basins of attraction exceed those of any classical Hopfield rule. We explain quantitatively this effect on uncorrelated random patterns. Moreover, we show that, for different structured datasets coming from computer science (random feature model, MNIST), physics (spin glasses) and biology (proteins), as the number of training examples increases the learned network goes beyond memorization, developing meaningful attractors with non-trivial correlations with test examples, thus showing the ability to generalize. Our results therefore reveal pseudo-likelihood works both as an efficient inference tool and as a principled mechanism for memory and generalization.

cond-mat.stat-mech↗

arXiv2025

Learning and Testing Inverse Statistical Problems For Interacting Systems Undergoing Phase Transition

Inverse problems arise in situations where data is available, but the underlying model is not. It can therefore be necessary to infer the parameters of the latter starting from the former. Statistical mechanics offers a toolbox of techniques to address this challenge. In this work, we illustrate three of the main methods: the Maximum Likelihood, Maximum Pseudo-Likelihood, and Mean-Field approaches. We begin with a thorough theoretical introduction to these methods, followed by their application to inference in several well-known statistical physics systems undergoing phase transitions. Namely, we consider the ordered and disordered Ising models, the vector Potts model, and the Blume-Capel model on both regular lattices and random graphs. This discussion is accompanied by a GitHub repository that allows users to both reproduce the results and experiment with new systems.

cond-mat.stat-mech↗

arXiv2025

Implicit bias produces neural scaling laws in learning curves, from perceptrons to deep networks

Scaling laws in deep learning -- empirical power-law relationships linking model performance to resource growth -- have emerged as simple yet striking regularities across architectures, datasets, and tasks. These laws are particularly impactful in guiding the design of state-of-the-art models, since they quantify the benefits of increasing data or model size, and hint at the foundations of interpretability in machine learning. However, most studies focus on asymptotic behavior at the end of training. In this work, we describe a richer picture by analyzing the entire training dynamics: we identify two novel \textit{dynamical} scaling laws that govern how performance evolves as function of different norm-based complexity measures. Combined, our new laws recover the well-known scaling for test error at convergence. Our findings are consistent across CNNs, ResNets, and Vision Transformers trained on MNIST, CIFAR-10 and CIFAR-100. Furthermore, we provide analytical support using a single-layer perceptron trained with logistic loss, where we derive the new dynamical scaling laws, and we explain them through the implicit bias induced by gradient-based training.

cs.LG↗

arXiv2024

Daydreaming Hopfield Networks and their surprising effectiveness on correlated data

To improve the storage capacity of the Hopfield model, we develop a version of the dreaming algorithm that perpetually reinforces the patterns to be stored (as in the Hebb rule), and erases the spurious memories (as in dreaming algorithms). For this reason, we called it Daydreaming. Daydreaming is not destructive and it converges asymptotically to stationary retrieval maps. When trained on random uncorrelated examples, the model shows optimal performance in terms of the size of the basins of attraction of stored examples and the quality of reconstruction. We also train the Daydreaming algorithm on correlated data obtained via the random-features model and argue that it spontaneously exploits the correlations thus increasing even further the storage capacity and the size of the basins of attraction. Moreover, the Daydreaming algorithm is also able to stabilize the features hidden in the data. Finally, we test Daydreaming on the MNIST dataset and show that it still works surprisingly well, producing attractors that are close to unseen examples and class prototypes.

cond-mat.dis-nn↗