Machine learning outperforms thermodynamics in measuring how well a many-body system learns a drive
نویسندگان
چکیده
Abstract Diverse many-body systems, from soap bubbles to suspensions polymers, learn and remember patterns in the drives that push them far equilibrium. This learning may be leveraged for computation, memory, engineering. Until now, has been detected with thermodynamic properties, such as work absorption strain. We progress beyond these macroscopic properties first defined equilibrium contexts: quantify statistical mechanical using representation learning, a machine-learning model which information squeezes through bottleneck. By calculating of bottleneck, we measure four facets systems’ learning: classification ability, memory capacity, discrimination novelty detection. Numerical simulations classical spin glass illustrate our technique. toolkit exposes self-organization eludes detection by measures: Our more reliably precisely detects quantifies matter while providing unifying framework learning.
منابع مشابه
Emergent thermodynamics in a quenched quantum many-body system.
We study the statistics of the work done, fluctuation relations, and irreversible entropy production in a quantum many-body system subject to the sudden quench of a control parameter. By treating the quench as a thermodynamic transformation we show that the emergence of irreversibility in the nonequilibrium dynamics of closed many-body quantum systems can be accurately characterized. We demonst...
متن کاملHow to Drive a B Machine
The B Method is a state based formal method that describes behaviour in terms of MACHINES whose states change under OPER ATIONS The process algebra CSP is an event based formalism that enables descriptions of patterns of system behaviour We present a com bination of the two views where a CSP process acts as a control executive and its events simply drive corresponding OPERATIONS We de ne con si...
متن کاملAssessing the Nonequilibrium Thermodynamics in a Quenched Quantum Many-Body System via Single Projective Measurements
متن کامل
Real-time Scheduling of a Flexible Manufacturing System using a Two-phase Machine Learning Algorithm
The static and analytic scheduling approach is very difficult to follow and is not always applicable in real-time. Most of the scheduling algorithms are designed to be established in offline environment. However, we are challenged with three characteristics in real cases: First, problem data of jobs are not known in advance. Second, most of the shop’s parameters tend to be stochastic. Third, th...
متن کاملMany-body localization in a quasiperiodic system
Shankar Iyer,1 Vadim Oganesyan,2,3,4 Gil Refael,1 and David A. Huse5 1Department of Physics, California Institute of Technology, MC 149-33, 1200 E. California Blvd., Pasadena, California 91125, USA 2Department of Engineering Science and Physics, College of Staten Island, CUNY, Staten Island, New York 10314, USA 3The Graduate Center, CUNY, 365 5th Ave., New York, New York 10016, USA 4KITP, UCSB,...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Scientific Reports
سال: 2021
ISSN: ['2045-2322']
DOI: https://doi.org/10.1038/s41598-021-88311-7