467 citations · 4.2k across the 83 of their papers we have counts for
18 papers · 1 filter
Restoring balance: principled under/oversampling of data for optimal classification
Emanuele Loffredo, Mauro Pastore, Simona Cocco +1
Class imbalance in real-world data poses a common bottleneck for machine learning tasks, since achieving good generalization on under-represented examples is often challenging. Mit…
Unlearning regularization for Boltzmann Machines
Enrico Ventura, Simona Cocco, Rémi Monasson +1
Boltzmann Machines (BMs) are graphical models with interconnected binary units, employed for the unsupervised modeling of data distributions. When trained on real data, BMs show th…
Information content in continuous attractor neural networks is preserved in the presence of moderate disordered background connectivity
Tobias Kühn, Rémi Monasson
Continuous attractor neural networks (CANN) form an appealing conceptual model for the storage of information in the brain. However a drawback of CANN is that they require finely t…
Barriers and Dynamical Paths in Alternating Gibbs Sampling of Restricted Boltzmann Machines
Clément Roussel, Simona Cocco, Rémi Monasson
Restricted Boltzmann Machines (RBM) are bi-layer neural networks used for the unsupervised learning of model distributions from data. The bipartite architecture of RBM naturally de…
On the Spectrum of Multi-Space Euclidean Random Matrices
Aldo Battista, Remi Monasson
We consider the additive superimposition of an extensive number of independent Euclidean Random Matrices in the high-density regime. The resolvent is computed with techniques from…
'Place-cell' emergence and learning of invariant data with restricted Boltzmann machines: breaking and dynamical restoration of continuous symmetries in the weight space
Moshir Harsh, Jérôme Tubiana, Simona Cocco +1
Distributions of data or sensory stimuli often enjoy underlying invariances. How and to what extent those symmetries are captured by unsupervised learning methods is a relevant que…
Capacity-resolution trade-off in the optimal learning of multiple low-dimensional manifolds by attractor neural networks
Aldo Battista, Rémi Monasson
Recurrent neural networks (RNN) are powerful tools to explain how attractors may emerge from noisy, high-dimensional dynamics. We study here how to learn the ~N^(2) pairwise intera…
Adaptive cluster expansion for the inverse Ising problem: convergence, algorithm and tests
Simona Cocco, Rémi Monasson
We present a procedure to solve the inverse Ising problem, that is to find the interactions between a set of binary variables from the measure of their equilibrium correlations. Th…
Small-correlation expansions for the inverse Ising problem
Vitor Sessak, Rémi Monasson
We present a systematic small-correlation expansion to solve the inverse Ising problem: find a set of couplings and fields corresponding to a given set of correlations and magnetiz…
Von Neumann's expanding model on random graphs
A De Martino, C Martelli, R Monasson +1
Within the framework of Von Neumann's expanding model, we study the maximum growth rate r achievable by an autocatalytic reaction network in which reactions involve a finite (fixed…
Inferring DNA sequences from mechanical unzipping: an ideal-case study
V. Baldazzi, S. Cocco, E. Marinari +1
We introduce and test a method to predict the sequence of DNA molecules from in silico unzipping experiments. The method is based on Bayesian inference and on the Viterbi decoding…
Exponentially hard problems are sometimes polynomial, a large deviation analysis of search algorithms for the random Satisfiability problem, and its application to stop-and-restart resolutions
S. Cocco, R. Monasson
A large deviation analysis of the solving complexity of random 3-Satisfiability instances slightly below threshold is presented. While finding a solution for such instances demands…
Trajectories in phase diagrams, growth processes and computational complexity: how search algorithms solve the 3-Satisfiability problem
Simona Cocco, Remi Monasson
Most decision and optimization problems encountered in practice fall into one of two categories with respect to any particular solving method or algorithm: either the problem is so…
From inherent structures to pure states: some simple remarks and examples
Giulio Biroli, Remi Monasson
The notions of pure states and inherent structures, i.e. stable configurations against 1-spin flip are discussed. We explain why these different concepts accidentally coincide in m…
2+p-SAT: Relation of Typical-Case Complexity to the Nature of the Phase Transition
R. Monasson, R. Zecchina, S. Kirkpatrick +2
Heuristic methods for solution of problems in the NP-Complete class of decision problems often reach exact solutions, but fail badly at "phase boundaries", across which the decisio…
Tricritical Points in Random Combinatorics: the (2+p)-SAT case
Remi Monasson, Riccardo Zecchina
The (2+p)-Satisfiability (SAT) problem interpolates between different classes of complexity theory and is believed to be of basic interest in understanding the onset of typical cas…
Optimisation problems and replica symmetry breaking in finite connectivity spin-glasses
Remi Monasson
A formalism capable of handling the first step of hierarchical replica symmetry breaking in finite-connectivity models is introduced. The emerging order parameter is claimed to be…
Statistical mechanics of the random K-SAT model
R. Monasson, R. Zecchina
The Random K-Satisfiability Problem, consisting in verifying the existence of an assignment of N Boolean variables that satisfy a set of M=alpha N random logical clauses containing…