Source author record

Franco Pellegrini

Franco Pellegrini appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

cond-mat.mes-hall cond-mat.mtrl-sci cond-mat.other cond-mat.dis-nn cond-mat.stat-mech Machine Learning physics.comp-ph quant-ph

Catalog footprint

What is connected

8works

8topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2020arXiv

An analytic theory of shallow networks dynamics for hinge loss classification

Neural networks have been shown to perform incredibly well in classification tasks over structured high-dimensional datasets. However, the learning dynamics of such networks is still poorly understood. In this paper we study in detail the training dynamics of a simple type of neural network: a single hidden layer trained to perform a classification task. We show that in a suitable mean-field limit this case maps to a single-node learning problem with a time-dependent dataset determined self-consistently from the average nodes population. We specialize our theory to the prototypical case of a linearly separable dataset and a linear hinge loss, for which the dynamics can be explicitly solved. This allow us to address in a simple setting several phenomena appearing in modern networks such as slowing down of training dynamics, crossover between rich and lazy learning, and overfitting. Finally, we asses the limitations of mean-field theory by studying the case of large but finite number of nodes and of training samples.

preprint2019arXiv

PANNA: Properties from Artificial Neural Network Architectures

Prediction of material properties from first principles is often a computationally expensive task. Recently, artificial neural networks and other machine learning approaches have been successfully employed to obtain accurate models at a low computational cost by leveraging existing example data. Here, we present a software package "Properties from Artificial Neural Network Architectures" (PANNA) that provides a comprehensive toolkit for creating neural network models for atomistic systems. Besides the core routines for neural network training, it includes data parser, descriptor builder and force-field generator suitable for integration within molecular dynamics packages. PANNA offers a variety of activation and cost functions, regularization methods, as well as the possibility of using fully-connected networks with custom size for each atomic species. PANNA benefits from the optimization and hardware-flexibility of the underlying TensorFlow engine which allows it to be used on multiple CPU/GPU/TPU systems, making it possible to develop and optimize neural network models based on large datasets.

preprint2016arXiv

CDW slips and giant frictional dissipation peaks at the NbSe$_2$ surface

Accessing, controlling and understanding nanoscale friction and dissipation is a crucial issue in nanotechnology, where moving elements are central. Recently, ultra-sensitive noncontact pendulum Atomic Force Microscope (AFM) succeeded in detecting the electronic friction drop caused by the onset of superconductivity in Nb, raising hopes that a wider variety of mechanisms of mechanical dissipation arising from electron organization into different collective phenomena will become accessible through this unconventional surface probe. Among them, the driven phase dynamics of charge-density-waves (CDWs) represents an outstanding challenge as a source of dissipation. Here we report a striking multiplet of AFM dissipation peaks arising at nanometer distances above the surface of NbSe$_2$ - a layered compound exhibiting an incommensurate CDW. Each peak appears at a well defined tip-surface interaction force of the order of a nN, and persists until T=70K where CDW short-range order is known to disappear. A theoretical model is presented showing that the peaks are connected to tip-induced local 2$π$ CDW phase slips. Under the attractive potential of the approaching tip, the local CDW surface phase landscape deforms continuously until a series of 2$π$ jumps occur between different values of the local phase. As the tip oscillates to and fro, each slip gives rise to a hysteresis cycle, appearing at a selected distance, the dissipation corresponding to "pumping" in and out a local slip in the surface CDW phase of NbSe$_2$.

preprint2016arXiv

Theory of charge density wave non-contact friction

A mechanism is proposed to describe the occurrence of distance-dependent dissipation peaks in the dynamics of an atomic force microscope tip oscillating over a surface characterized by a charge density wave state. The dissipation has its origin in the hysteretic behavior of the tip oscillations occurring at positions compatible with a localized phase slip of the charge density wave. This model is supported through static and dynamic numerical simulations of the tip surface interaction and is in good qualitative agreement with recently performed experiments on a NbSe$_2$ sample.

preprint2014arXiv

Charge density wave surface phase slips and non-contact nanofriction

Bulk electrical dissipation caused by charge-density-wave (CDW) depinning and sliding is a classic subject. We present a novel local, nanoscale mechanism describing the occurrence of mechanical dissipation peaks in the dynamics of an atomic force microscope tip oscillating above the surface of a CDW material. Local surface 2$π$ slips of the CDW phase are predicted to take place giving rise to mechanical hysteresis and large dissipation at discrete tip surface distances. The results of our static and dynamic numerical simulations are believed to be relevant to recent experiments on NbSe$_2$; other candidate systems in which similar effects should be observable are also discussed.

preprint2014arXiv

Non-contact dissipation on the surface of charge density wave materials

preprint2011arXiv

Crossover from adiabatic to antiadiabatic quantum pumping with dissipation

Quantum pumping, in its different forms, is attracting attention from different fields, from fundamental quantum mechanics, to nanotechnology, to superconductivity. We investigate the crossover of quantum pumping from the adiabatic to the anti-adiabatic regime in the presence of dissipation, and find general and explicit analytical expressions for the pumped current in a minimal model describing a system with the topology of a ring forced by a periodic modulation of frequency omega. The solution allows following in a transparent way the evolution of pumped DC current from much smaller to much larger omega values than the other relevant energy scale, the energy splitting introduced by the modulation. We find and characterize a temperature-dependent optimal value of the frequency for which the pumped current is maximal.

preprint2010arXiv

Atomic spin sensitive dissipation on magnetic surfaces

We identify the mechanism of energy dissipation relevant to spin-sensitive nanomechanics including the recently introduced magnetic exchange force microscopy, where oscillating magnetic tips approach surface atomic spins. The tip-surface exchange couples spin and atom coordinates, leading to a spin-phonon problem with Caldeira-Leggett type dissipation. In the overdamped regime, that can lead to a hysteretic flip of the local spin with a large spin-dependent dissipation, even down to the very low experimental tip oscillation frequencies, describing recent observations for Fe tips on NiO. A phase transition to an underdamped regime with dramatic drop of magnetic tip dissipation should in principle be possible by tuning tip-surface distance.