Marco Leonetti

Marco Leonetti

Auto-generated profile, based on publicly available information

Marco Leonetti

Center for Life Nano- & Neuro-Science, Italian Institute of Technology, Rome, Italy

Research Focus & Metrics

Academic Career Length

19Years (2008-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 September 2011 in Nature Photonics
Verified
Journal Article
Closed access
PDF file
MLMarco Leonetti
Marco Leonetti
CCClaudio Conti
Claudio Conti
CLCefe López
Cefe López

30 September 2008 in Physical Review Letters
Verified
Journal Article
Closed access
PDF file
CCClaudio Conti
Claudio Conti
MLM. Leonetti
M. Leonetti
AFAndrea Fratalocchi
Andrea Fratalocchi
LALuca Angelani
Luca Angelani
Giancarlo RuoccoGiancarlo Ruocco
Giancarlo Ruocco
The complex processes underlying the generation of a coherent emission from the multiple scattering of photons and wave localization in the presence of structural disorder are still mostly unexplored. Here we show that a single nonlinear Schrödinger equation, playing the role of the Schwalow-Townes law for standard lasers, quantitatively reproduces experimental results and three-dimensional time-domain parallel simulations of a colloidal laser system.

18 February 2013 in Applied Physics Letters
Verified
Journal Article
Closed access
PDF file
MLMarco Leonetti
Marco Leonetti
CLCefe López
Cefe López
We demonstrate an experimental technique that allows to achieve a robust control on the emission spectrum of a micro random laser and to select individual modes with sub-nanometer resolution. The presented approach relies on an optimization protocol of the spatial profile of the pump beam. Here we demonstrate not only the possibility to increase the emission at a wavelength but also that we can “isolate” an individual peak suppressing unwanted contributions form other modes.

10 January 2017 in Frontiers in Computational Neuroscience
Verified
Journal Article
Open access
PDF file
Viola FolliViola Folli
Viola Folli
MLMarco Leonetti
Marco Leonetti
Giancarlo RuoccoGiancarlo Ruocco
Giancarlo Ruocco
Recurrent neural networks (RNN) have traditionally been of great interest for their capacity to store memories. In past years, several works have been devoted to determine the maximum storage capacity of RNN, especially for the case of the Hopfield network, the most popular kind of RNN. Analyzing the thermodynamic limit of the statistical properties of the Hamiltonian corresponding to the Hopfield neural network, it has been shown in the literature that the retrieval errors diverge when the number of stored memory patterns ($P$) exceeds a fraction ( $\approx 14\%$) of the network size $N$. In this paper, we study the storage performance of a generalized Hopfield model, where the diagonal elements of the connection matrix are allowed to be different from zero. We investigate this model at finite $N$. We give an analytical expression for the number of retrieval errors and show that, by increasing the number of stored patterns over a certain threshold, the errors start to decrease and reach values below unit for $P \gg N$. We demonstrate that the strongest trade-off between efficiency and effectiveness relies on the number of patterns ($P$) that are stored in the network by appropriately fixing the connection weights. When $P$$\gg$$N$ and the diagonal elements of the adjacency matrix are not forced to be zero, the optimal storage capacity is obtained with a number of stored memories much larger than previously reported. This theory paves the way to the design of RNN with high storage capacity and able to retrieve the desired pattern without distortions.

21 May 2021 in Proceedings of the National Academy of Sciences
Verified
Journal Article
Closed access
PDF file
MLM. Leonetti
M. Leonetti
EHE. Hörmann
E. Hörmann
LLL. Leuzzi
L. Leuzzi
GPGiorgio Parisi
Giorgio Parisi
Giancarlo RuoccoGiancarlo Ruocco
Giancarlo Ruocco
Spin glasses (SGs) are paradigmatic models for physical, computer science, biological, and social systems. The problem of studying the dynamics for SG models is nondetermistic polynomial-time (NP) hard; that is, no algorithm solves it in polynomial time. Here we implement the optical simulation of an SG, exploiting the N segments of a wavefront-shaping device to play the role of the spin variables, combining the interference downstream of a scattering material to implement the random couplings between the spins (the Jij matrix) and measuring the light intensity on a number P of targets to retrieve the energy of the system. By implementing a plain Metropolis algorithm, we are able to simulate the spin model dynamics, while the degree of complexity of the potential energy landscape and the region of phase diagram explored are user defined, acting on the ratio P/N=α. We study experimentally, numerically, and analytically this Hopfield-like system displaying a paramagnetic, ferromagnetic, and SG phase, and we demonstrate that the transition temperature Tg to the glassy phase from the paramagnetic phase grows with α. We demonstrate the computational advantage of the optical SG where interaction terms are realized simultaneously when the independent light rays interfere on the detector’s surface. This inherently parallel measurement of the energy provides a speedup with respect to purely in silico simulations scaling with N. Significance Optics is a unique platform to speed up computation, providing advantages especially for inherently parallel operations. Here we demonstrate an optical experiment enabling the simulation of a paradigmatic prototype model for statistical mechanics: the spin glass (SG). SGs are systems presenting all the aspects of complexity, such as many equilibrium states producing multiple relaxation times and nontrivial phenomenology. With the optical approach, the spin state is mapped on the mirrors of a fast adaptive optics device illuminated by a laser, and the energy calculation is performed by collecting the intensity after all the controlled light rays interfered in strongly scattering medium. The optical simulation enables beating the digital calculation in terms of speed for large-sized SG.

14 May 2014 in Physical Review Letters
Verified
Journal Article
Closed access
PDF file
MLMarco Leonetti
Marco Leonetti
SKSalman Karbasi
Salman Karbasi
AMArash Mafi
Arash Mafi
CCClaudio Conti
Claudio Conti
We report the experimental observation of the interaction and attraction of many localized modes in a two-dimensional system realized by a disordered optical fiber supporting transverse Anderson localization. We show that a nonlocal optically nonlinear response of thermal origin alters the localization length by an amount determined by the optical power and also induces an action at a distance between the localized modes and their spatial migration. Evidence of a collective and strongly interacting regime is given.