arXiv subjects

Luca Gammaitoni

Publications and source records attributed to Luca Gammaitoni.

9 recordsLinked to original sources

arXiv2024

Prediction and Inference: From Models and Data to Artificial Intelligence

In this paper we present a discussion of the basic aspects of the well-known problem of prediction and inference in physics, with specific attention to the role of models, the use of data and the application of recent developments in artificial intelligence. By focussing in the time evolution of dynamic system, it is shown that main difficulties in predictions arise due to the presence of few factors as: the occurrence of chaotic dynamics, the existence of many variables with very different characteristic time-scales and the lack of an accurate understanding of the underlying physical phenomena. It is shown that a crucial role is assigned to the preliminary identification of the proper variables, their selection and the identification of an appropriate level of description (coarse-graining procedure).

physics.gen-ph↗

arXiv2018

The superconductor-superinsulator transition: S-duality and QCD on the desktop

We show that the nature of quantum phases around the superconductor-insulator transition (SIT) is controlled by charge-vortex topological interactions and does not depend on the details of material parameters and disorder. We find three distinct phases, superconductor, superinsulator and bosonic topological insulator. The superinsulator is a state of matter with infinite resistance in a finite temperature range, which is the S-dual of the superconductor and in which charge transport is prevented by electric strings binding charges of opposite sign. The electric strings ensuring linear confinement of charges are generated by instantons and are dual to superconducting Abrikosov vortices. Material parameters and disorder enter the London penetration depth of the superconductor, the string tension of the superinsulator and the quantum fluctuation parameter driving the transition between them. They are entirely encoded in four phenomenological parameters of a topological gauge theory of the SIT. Finally, we point out that, in the context of strong coupling gauge theories, the many-body localization phenomenon that is often referred to as an underlying mechanism for superinsulation is a mere transcription of the well-known phenomenon of confinement into solid state physics language and is entirely driven by endogenous disorder embodied by instantons with no need of exogenous disorder.

cond-mat.supr-con↗

arXiv2018

Vogel-Fulcher-Tamman criticality of 3D superinsulators

It has been believed that the superinsulating state which is the low-temperature charge Berezinskii-Kosterlitz- Thouless (BKT) phase can exist only in two dimensions. We develop a general gauge description of the su- perinsulating state and the related deconfinement transition of Cooper pairs and predict the existence of the superinsulating state in three dimensions (3d). We find that 3d superinsulators exhibit Vogel-Fulcher-Tammann (VFT) critical behavior at the phase transition. This is the 3d string analogue of the Berezinski-Kosterlitz- Thouless (BKT) criticality for logarithmically and linearly interacting point particles in 2d. Our results show that singular exponential scaling behaviors of the BKT type are generic for phase transitions associated with the condensation of topological excitations.

cond-mat.str-el↗

arXiv2017

Operating Gravitational Wave Detectors far from equilibrium

Tiny vibrations of mechanical structures are the main limiting cause in a number of high sensitivity measurement apparatus, chief among them the most sensitive displacement apparatus on earth: gravitational wave interferometers. Such devices are usually operated at equilibrium and small fluctuations are perceived as noise that sets a lower limit to the detection capabilities. An example is the so-called thermal noise, ubiquitous and unavoidable. In this letter we present an approach aimed at operating the interferometer out of equilibrium. We show that selective cooling of single modes of the mechanical structure is able to positively impact the measurement sensitivity, in selected frequency ranges. Experiments conducted on thin silica membranes show promising results for the implementation of such technique in next generation gravitational wave detectors.

physics.ins-det↗

arXiv2017

Cost of remembering a bit of information

In 1961, Rolf Landauer pointed out that resetting a binary memory requires a minimum energy of $k_BT \ln(2)$. However, once written, any memory is doomed to loose its content if no action is taken. To avoid memory losses, a refresh procedure is periodically performed. In this paper we present a theoretical model and an experiment on a micro-electro-mechanical system to evaluate the minimum energy required to preserve one bit of information over time. Two main conclusions are drawn: i) in principle the energetic cost to preserve information for a fixed time duration with a given error probability can be arbitrarily reduced if the refresh procedure is performed often enough; ii) the Heisenberg uncertainty principle sets an upper bound on the memory lifetime.

cond-mat.stat-mech↗

arXiv2016

Landauer Bound for Analog Computing Systems

By establishing a relation between information erasure and continuous phase transitions we generalise the Landauer bound to analog computing systems. The entropy production per degree of freedom during erasure of an analog variable (reset to standard value) is given by the logarithm of the configurational volume measured in units of its minimal quantum. As a consequence every computation has to be carried on with a finite number of bits and infinite precision is forbidden by the fundamental laws of physics, since it would require an infinite amount of energy.

cs.IT↗

arXiv2016

Sub $k_B T$ micro-electromechanical irreversible logic gate

In modern computers, computation is performed by assembling together sets of logic gates. Popular gates like AND, OR, XOR, processing two logic inputs and yielding one logic output, are often addressed as irreversible logic gates where the sole knowledge of the output logic value, is not sufficient to infer the logic value of the two inputs. Such gates are usually believed to be bounded to dissipate a finite minimum amount of energy determined by the input-output information difference. Here we show that this is not necessarily the case, by presenting an experiment where a OR logic gate, realized with a micro electromechanical cantilever, is operated with energy well below the expected limit, provided the operation is slow enough and frictional phenomena are properly addressed.

cond-mat.mes-hall↗

arXiv2011

Beating the Landauer's limit by trading energy with uncertainty

According to the International Technology Roadmap for Semiconductors in the next 10-15 years the limits imposed by the physics of switch operation will be the major roadblock for future scaling of the CMOS technology. Among these limits the most fundamental is represented by the so-called Shannon-von Neumann-Landauer limit that sets a lower bound to the minimum heat dissipated per bit erasing operation. Here we show that in a nanoscale switch, operated at finite temperature T, this limit can be beaten by trading the dissipated energy with the uncertainty in the distinguishability of switch logic states. We establish a general relation between the minimum required energy and the maximum error rate in the switch operation and briefly discuss the potential applications in the design of future switches.

cond-mat.mtrl-sci↗

arXiv2007

Noise Limited Computational Speed

In modern transistor based logic gates, the impact of noise on computation has become increasingly relevant since the voltage scaling strategy, aimed at decreasing the dissipated power, has increased the probability of error due to the reduced switching threshold voltages. In this paper we discuss the role of noise in a two state model that mimic the dynamics of standard logic gates and show that the presence of the noise sets a fundamental limit to the computing speed. An optimal idle time interval that minimizes the error probability, is derived.

cs.AR↗