Showing 1–12 of 12 results for author: Rocchi, J

On the number of limit cycles in diluted neural networks

Authors: Sungmin Hwang, Enrico Lanza, Giorgio Parisi, Jacopo Rocchi, Giancarlo Ruocco, Francesco Zamponi

Abstract: We consider the storage properties of temporal patterns, i.e. cycles of finite lengths, in neural networks represented by (generally asymmetric) spin glasses defined on random graphs. Inspired by the observation that dynamics on sparse systems have more basins of attractions than the dynamics of densely connected ones, we consider the attractors of a greedy dynamics in sparse topologies, considere… ▽ More We consider the storage properties of temporal patterns, i.e. cycles of finite lengths, in neural networks represented by (generally asymmetric) spin glasses defined on random graphs. Inspired by the observation that dynamics on sparse systems have more basins of attractions than the dynamics of densely connected ones, we consider the attractors of a greedy dynamics in sparse topologies, considered as proxy for the stored memories. We enumerate them using numerical simulation and extend the analysis to large systems sizes using belief propagation. We find that the logarithm of the number of such cycles is a non monotonic function of the mean connectivity and we discuss the similarities with biological neural networks describing the memory capacity of the hippocampus. △ Less

Submitted 1 January, 2020; originally announced January 2020.

Comments: 10 pages, 11 figures

Journal ref: Journal of Statistical Physics 181, 2304-2321 (2020)

Large deviations of glassy effective potentials

Authors: Silvio Franz, Jacopo Rocchi

Abstract: The theory of glassy fluctuations can be formulated in terms of disordered effective potentials. While the properties of the average potentials are well understood, the study of the fluctuations has been so far quite limited. Close to the MCT transition, fluctuations induced by the dynamical heterogeneities in supercooled liquids can be described by a cubic field theory in presence of a random fie… ▽ More The theory of glassy fluctuations can be formulated in terms of disordered effective potentials. While the properties of the average potentials are well understood, the study of the fluctuations has been so far quite limited. Close to the MCT transition, fluctuations induced by the dynamical heterogeneities in supercooled liquids can be described by a cubic field theory in presence of a random field term. In this paper we set up the general problem of the large deviations going beyond the assumption of the vicinity to $T_{MCT}$ and analyze it in the paradigmatic case of spherical ($p$-spin) glass models. This tool can be applied to study the probability of the observation of a dynamics with memory of the initial condition in regimes where, typically, the correlation $C(t,0)$ decays to zero at long times, at finite $T$ and at $T=0$. △ Less

Submitted 28 September, 2019; originally announced September 2019.

Journal ref: J. Phys. A: Math. Theo. (2020)

Inverse problems for structured datasets using parallel TAP equations and RBM

Authors: Aurélien Decelle, Sungmin Hwang, Jacopo Rocchi, Daniele Tantari

Abstract: We propose an efficient algorithm to solve inverse problems in the presence of binary clustered datasets. We consider the paradigmatic Hopfield model in a teacher student scenario, where this situation is found in the retrieval phase. This problem has been widely analyzed through various methods such as mean-field approaches or the pseudo-likelihood optimization. Our approach is based on the estim… ▽ More We propose an efficient algorithm to solve inverse problems in the presence of binary clustered datasets. We consider the paradigmatic Hopfield model in a teacher student scenario, where this situation is found in the retrieval phase. This problem has been widely analyzed through various methods such as mean-field approaches or the pseudo-likelihood optimization. Our approach is based on the estimation of the posterior using the Thouless-Anderson-Palmer (TAP) equations in a parallel updating scheme. At the difference with other methods, it allows to retrieve the exact patterns of the teacher and the parallel update makes it possible to apply it for large system sizes. We also observe that the Approximate Message Passing (AMP) equations do not reproduce the expected behavior in the direct problem, questioning the standard practice used to obtain time indexes coming from Belief Propagation (BP). We tackle the same problem using a Restricted Boltzmann Machine (RBM) and discuss the analogies between the two algorithms. △ Less

Submitted 27 June, 2019; originally announced June 2019.

Journal ref: Sci Rep 11, 19990 (2021)

A fast and accurate algorithm for inferring sparse Ising models via parameters activation to maximize the pseudo-likelihood

Authors: Silvio Franz, Federico Ricci-Tersenghi, Jacopo Rocchi

Abstract: We propose a new algorithm to learn the network of the interactions of pairwise Ising models. The algorithm is based on the pseudo-likelihood method (PLM), that has already been proven to efficiently solve the problem in a large variety of cases. Our present implementation is particularly suitable to address the case of sparse underlying topologies and it is based on a careful search of the most i… ▽ More We propose a new algorithm to learn the network of the interactions of pairwise Ising models. The algorithm is based on the pseudo-likelihood method (PLM), that has already been proven to efficiently solve the problem in a large variety of cases. Our present implementation is particularly suitable to address the case of sparse underlying topologies and it is based on a careful search of the most important parameters in their high dimensional space. We call this algorithm Parameters Activation to Maximize Pseudo-Likelihood (PAMPL). Numerical tests have been performed on a wide class of models such as random graphs and finite dimensional lattices with different type of couplings, both ferromagnetic and spin glasses. These tests show that PAMPL improves the performances of the fastest existing algorithms. △ Less

Submitted 17 February, 2019; v1 submitted 31 January, 2019; originally announced January 2019.

Slow Spin Dynamics and Self-Sustained Clusters in Sparsely Connected Systems

Authors: Jacopo Rocchi, David Saad, Chi Ho Yeung

Abstract: To identify emerging microscopic structures in low temperature spin glasses, we study self-sustained clusters (SSC) in spin models defined on sparse random graphs. A message-passing algorithm is developed to determine the probability of individual spins to belong to SSC. Results for specific instances, which compare the predicted SSC associations with the dynamical properties of spins obtained fro… ▽ More To identify emerging microscopic structures in low temperature spin glasses, we study self-sustained clusters (SSC) in spin models defined on sparse random graphs. A message-passing algorithm is developed to determine the probability of individual spins to belong to SSC. Results for specific instances, which compare the predicted SSC associations with the dynamical properties of spins obtained from numerical simulations, show that SSC association identifies individual slow-evolving spins. This insight gives rise to a powerful approach for predicting individual spin dynamics from a single snapshot of an equilibrium spin configuration, namely from limited static information, which can be used to devise generic prediction tools applicable to a wide range of areas. △ Less

Submitted 4 June, 2017; originally announced June 2017.

Journal ref: Phys. Rev. E 97, 062154 (2018)

High storage capacity in the Hopfield model with auto-interactions - stability analysis

Authors: Jacopo Rocchi, David Saad, Daniele Tantari

Abstract: Recent studies point to the potential storage of a large number of patterns in the celebrated Hopfield associative memory model, well beyond the limits obtained previously. We investigate the properties of new fixed points to discover that they exhibit instabilities for small perturbations and are therefore of limited value as associative memories. Moreover, a large deviations approach also shows… ▽ More Recent studies point to the potential storage of a large number of patterns in the celebrated Hopfield associative memory model, well beyond the limits obtained previously. We investigate the properties of new fixed points to discover that they exhibit instabilities for small perturbations and are therefore of limited value as associative memories. Moreover, a large deviations approach also shows that errors introduced to the original patterns induce additional errors and increased corruption with respect to the stored patterns. △ Less

Submitted 25 April, 2017; originally announced April 2017.

Self-Sustained Clusters as Drivers of Computational Hardness in $p$-spin Models

Authors: Jacopo Rocchi, David Saad, Chi Ho Yeung

Abstract: While macroscopic properties of spin glasses have been thoroughly investigated, their manifestation in the corresponding microscopic configurations is much less understood. Cases where both descriptions have been provided, such as constraint satisfaction problems, are limited to their ground state properties. To identify the emerging microscopic structures with macroscopic phases at different temp… ▽ More While macroscopic properties of spin glasses have been thoroughly investigated, their manifestation in the corresponding microscopic configurations is much less understood. Cases where both descriptions have been provided, such as constraint satisfaction problems, are limited to their ground state properties. To identify the emerging microscopic structures with macroscopic phases at different temperatures, we study the $p$-spin model with $p\!=\!3$. We investigate the properties of self-sustained clusters, defined as variable sets where in-cluster induced fields dominate over the field induced by out-cluster spins, giving rise to stable configurations with respect to fluctuations. We compute the entropy of self-sustained clusters as a function of temperature and their sizes. In-cluster fields properties and the difference between in-cluster and out-cluster fields support the observation of slow-evolving spins in spin models. The findings are corroborated by observations in finite dimensional lattices at low temperatures. △ Less

Submitted 9 November, 2016; originally announced November 2016.

Journal ref: Phys. Rev. B 96, 024415 (2017)

Emerging interdependence between stock values during financial crashes

Authors: Jacopo Rocchi, Enoch Yan Lok Tsui, David Saad

Abstract: To identify emerging interdependencies between traded stocks we investigate the behavior of the stocks of FTSE 100 companies in the period 2000-2015, by looking at daily stock values. Exploiting the power of information theoretical measures to extract direct influences between multiple time series, we compute the information flow across stock values to identify several different regimes. While sma… ▽ More To identify emerging interdependencies between traded stocks we investigate the behavior of the stocks of FTSE 100 companies in the period 2000-2015, by looking at daily stock values. Exploiting the power of information theoretical measures to extract direct influences between multiple time series, we compute the information flow across stock values to identify several different regimes. While small information flows is detected in most of the period, a dramatically different situation occurs in the proximity of global financial crises, where stock values exhibit strong and substantial interdependence for a prolonged period. This behavior is consistent with what one would generally expect from a complex system near criticality in physical systems, showing the long lasting effects of crashes on stock markets. △ Less

Submitted 8 November, 2016; originally announced November 2016.

Cross correlations of the American baby names

Authors: Paolo Barucca, Jacopo Rocchi, Enzo Marinari, Giorgio Parisi, Federico Ricci-Tersenghi

Abstract: The quantitative description of cultural evolution is a challenging task. The most difficult part of the problem is probably to find the appropriate measurable quantities that can make more quantitative such evasive concepts as, for example, dynamics of cultural movements, behavior patterns and traditions of the people. A strategy to tackle this issue is to observe particular features of human act… ▽ More The quantitative description of cultural evolution is a challenging task. The most difficult part of the problem is probably to find the appropriate measurable quantities that can make more quantitative such evasive concepts as, for example, dynamics of cultural movements, behavior patterns and traditions of the people. A strategy to tackle this issue is to observe particular features of human activities, i.e. cultural traits, such as names given to newborns. We study the names of babies born in the United States of America from 1910 to 2012. Our analysis shows that groups of different correlated states naturally emerge in different epochs, and we are able to follow and decrypt their evolution. While these groups of states are stable across many decades, a sudden reorganization occurs in the last part of the twentieth century. We think that this kind of quantitative analysis can be possibly extended to other cultural traits: although databases covering more than one century (as the one we used) are rare, the cultural evolution on shorter time scales can be studied thanks to the fact that many human activities are usually recorded in the present digital era. △ Less

Submitted 13 May, 2015; v1 submitted 10 October, 2014; originally announced October 2014.

Comments: submitted for consideration to PNAS

Journal ref: PNAS 112, 7943-7947 (2015)

Statistical mechanics of the spherical hierarchical model with random fields

Authors: Fernando L. Metz, Jacopo Rocchi, Pierfrancesco Urbani

Abstract: We study analytically the equilibrium properties of the spherical hierarchical model in the presence of random fields. The expression for the critical line separating a paramagnetic from a ferromagnetic phase is derived. The critical exponents characterising this phase transition are computed analytically and compared with those of the corresponding $D$-dimensional short-range model, leading to co… ▽ More We study analytically the equilibrium properties of the spherical hierarchical model in the presence of random fields. The expression for the critical line separating a paramagnetic from a ferromagnetic phase is derived. The critical exponents characterising this phase transition are computed analytically and compared with those of the corresponding $D$-dimensional short-range model, leading to conclude that the usual map** between one dimensional long-range models and $D$-dimensional short-range models holds exactly for this system, in contrast to models with Ising spins. Moreover, the critical exponents of the pure model and those of the random field model satisfy a relationship that mimics the dimensional reduction rule. The absence of a spin-glass phase is strongly supported by the local stability analysis of the replica symmetric saddle-point as well as by an independent computation of the free-energy using a renormalization-like approach. This latter result enlarges the class of random field models for which the spin-glass phase has been recently ruled out. △ Less

Submitted 5 June, 2014; originally announced June 2014.

Comments: 23 pages, 2 figures

Journal ref: J. Stat. Mech. P09018 (2014)

Ensemble renormalization group for the random field hierarchical model

Authors: Aurélien Decelle, Giorgio Parisi, Jacopo Rocchi

Abstract: The Renormalization Group (RG) methods are still far from being completely understood in quenched disordered systems. In order to gain insight into the nature of the phase transition of these systems, it is common to investigate simple models. In this work we study a real-space RG transformation on the Dyson hierarchical lattice with a random field, which led to a reconstruction of the RG flow and… ▽ More The Renormalization Group (RG) methods are still far from being completely understood in quenched disordered systems. In order to gain insight into the nature of the phase transition of these systems, it is common to investigate simple models. In this work we study a real-space RG transformation on the Dyson hierarchical lattice with a random field, which led to a reconstruction of the RG flow and to an evaluation of the critical exponents of the model at T = 0. We show that this method gives very accurate estimations of the critical exponents, by comparing our results with the ones obtained by some of us using an independent method. △ Less

Submitted 8 January, 2014; originally announced January 2014.

Journal ref: Phys. Rev. E 89, 032132 (2014)

Critical Exponents of the Random Field Hierarchical Model

Authors: Giorgio Parisi, Jacopo Rocchi

Abstract: We have studied the one dimensional Dyson hierarchical model in presence of a random field. This is a long range model where the interactions scale with the distance with a power law-like form J(r) ~ r^{-ρ} and we can explore mean field and non-mean field behavior by changing ρ. Thus, it can be used to approach the phase transitions in finite-dimensional disordered models. We studied the model at… ▽ More We have studied the one dimensional Dyson hierarchical model in presence of a random field. This is a long range model where the interactions scale with the distance with a power law-like form J(r) ~ r^{-ρ} and we can explore mean field and non-mean field behavior by changing ρ. Thus, it can be used to approach the phase transitions in finite-dimensional disordered models. We studied the model at T=0 and we numerically computed its critical exponents in the non-mean field region for Gaussian disorder. We then computed an analytic expression for the critical exponent δ, that holds in the non-mean field region, and we noted an interesting relation between the critical exponents of the disordered model and the ones of the pure model, that seems to break down in the non-mean field region. We finally compare our results for the critical exponents with the expected ones in D-dimensional short range models and with the ones of the straightforward one dimensional long range model. △ Less

Submitted 28 September, 2013; originally announced September 2013.

Comments: 9 pages, 3 figures

Journal ref: Phys. Rev. B 90, 024203 (2014)