English
Related papers

Related papers: Finite size scaling in neural networks

200 papers

For systems with infinite-order phase transitions, in which an order parameter smoothly becomes nonzero, a new observable for finite-size scaling analysis is suggested. By construction this new observable has the favourable property of…

Statistical Mechanics · Physics 2016-09-15 Rick Keesman , Jules Lamers , R. A. Duine , G. T. Barkema

The information processing abilities of a multilayer neural network with a number of hidden units scaling as the input dimension are studied using statistical mechanics methods. The mapping from the input layer to the hidden units is…

Statistical Mechanics · Physics 2009-11-07 Michal Rosen-Zvi , Andreas Engel , Ido Kanter

To study the kinetics of phase separation in active matter systems, we consider models that impose a Vicsek-type self-propulsion rule on otherwise passive particles interacting via the Lennard-Jones potential. Two types of kinetics are of…

Statistical Mechanics · Physics 2023-10-31 Tanay Paul , Nalina Vadakkayil , Subir K. Das

Pinning models are built from discrete renewal sequences by rewarding (or penalizing) the trajectories according to their number of renewal epochs up to time $N$, and $N$ is then sent to infinity. They are statistical mechanics models to…

Probability · Mathematics 2015-02-27 Julien Sohier

Single-time and two-time correlators are computed exactly in the $1D$ Glauber-Ising model after a quench to zero temperature and on a periodic chain of finite length $N$, using a simple analytical continuation technique. Besides the general…

Statistical Mechanics · Physics 2025-01-30 Malte Henkel

The scaling exponent and scaling function for the 1D single species coagulation model $(A+A\rightarrow A)$ are shown to be universal, i.e. they are not influenced by the value of the coagulation rate. They are independent of the initial…

Condensed Matter · Physics 2009-10-22 Klaus Krebs , Markus Pfannmueller , Horatiu Simon , Birgit Wehefritz

In a class of methods for measurement of available path capacity and other capacity-related metrics in a network, trains of probe packets are transmitted from a sender to a receiver across a network path, and the sequences of time stamps at…

Networking and Internet Architecture · Computer Science 2015-05-26 Svante Ekelin , Andreas Johnsson , Christofer Flinta

We investigate finite size scaling in percolating widthless stick systems with variable aspect ratios in an extensive Monte Carlo simulation study. A generalized scaling function is introduced to describe the scaling behavior of the…

Soft Condensed Matter · Physics 2012-02-03 Milan Žeželj , Igor Stanković , Aleksandar Belić

We address the problem of communicating domain knowledge from a user to the designer of a clustering algorithm. We propose a protocol in which the user provides a clustering of a relatively small random sample of a data set. The algorithm…

Machine Learning · Statistics 2015-06-22 Hassan Ashtiani , Shai Ben-David

The storage capacity of the Hopfield model is about 15% of the network size. It can be increased significantly in the Potts-glass model of the associative memory only. In this model neurons can be in more than two different states. We show…

Disordered Systems and Neural Networks · Physics 2007-05-23 B. V. Kryzhanovsky , L. B. Litinskii , A. L. Mikaelyan

We introduce two practical properties of hierarchical clustering methods for (possibly asymmetric) network data: excisiveness and linear scale preservation. The latter enforces imperviousness to change in units of measure whereas the former…

Machine Learning · Computer Science 2016-07-22 Gunnar Carlsson , Facundo Mémoli , Alejandro Ribeiro , Santiago Segarra

A long standing open problem in the theory of neural networks is the development of quantitative methods to estimate and compare the capabilities of different architectures. Here we define the capacity of an architecture by the binary…

Machine Learning · Computer Science 2019-03-29 Pierre Baldi , Roman Vershynin

We study the large-$r$ behavior of the bulk order-parameter correlation function $G(\bf{r})$ for $T>T_c$ within the lattice $\phi^4$ theory. We also study the large-$L$ behavior of the susceptibility $\chi$ of the confined lattice system of…

Statistical Mechanics · Physics 2009-10-31 X. S. Chen , V. Dohm

The network scale-up method enables researchers to estimate the size of hidden populations, such as drug injectors and sex workers, using sampled social network data. The basic scale-up estimator offers advantages over other size estimation…

Applications · Statistics 2016-11-14 Dennis M. Feehan , Matthew J. Salganik

Identifying the brain's neuronal cluster size to be presented as nodes in a network computation is critical to both neuroscience and artificial intelligence, as these define the cognitive blocks required for building intelligent…

Neurons and Cognition · Quantitative Biology 2020-02-26 Anar Amgalan , Patrick Taylor , Lilianne R. Mujica-Parodi , Hava T. Siegelmann

Many machine learning models require setting a parameter that controls their size before training, e.g. number of neurons in DNNs, or inducing points in GPs. Increasing capacity typically improves performance until all the information from…

Machine Learning · Statistics 2025-12-22 Guiomar Pescador-Barrios , Sarah Filippi , Mark van der Wilk

Finite-size scaling (FSS) is applied to net-baryon cumulant ratios $C_2/C_1$, $C_3/C_2$, $C_4/C_2$, $C_3/C_1$, and $C_4/C_1$ measured in Au+Au collisions over the Beam Energy Scan Phase~I range $\sqrt{s_{NN}}=7.7$--$200$~GeV to constrain…

Nuclear Experiment · Physics 2026-03-31 Roy A. Lacey

We calculate universal finite size scaling functions for the order parameter and the longitudinal susceptibility of the three-dimensional O(4) model. The phase transition of this model is supposed to be in the same universality class as the…

High Energy Physics - Lattice · Physics 2015-06-18 J. Engels , F. Karsch

Subspace clustering algorithms are notorious for their scalability issues because building and processing large affinity matrices are demanding. In this paper, we introduce a method that simultaneously learns an embedding space along…

Computer Vision and Pattern Recognition · Computer Science 2018-11-06 Tong Zhang , Pan Ji , Mehrtash Harandi , Richard Hartley , Ian Reid

Diffusion on a T fractal lattice under the influence of topological biasing fields is studied by finite size scaling methods. This allows to avoid proliferation and singularities which would arise in a renormalization group approach on…

Condensed Matter · Physics 2015-06-25 G. Sartoni , A. L. Stella
‹ Prev 1 3 4 5 6 7 10 Next ›