Simulated annealing and Boltzmann machines

By E. Aarts, J. Korst. 2003

1,267 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/4701df2e8e301157844d8eb17ec500af5f476d96

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Super-convergence: very fast training of neural networks using large learning rates (2018, background): “, 2009] and simulated annealing [Aarts and Korst, 1988], both of which have a long history of use in deep learning.”
Flash-flood hazard assessment using ensembles and Bayesian-based machine learning models: Application of the simulated annealing feature selection method. (2019, methodology): “SA is an efficient 233 global optimization method based on a random search technique, widely used to identify the 234 optimum in relatively large design space (Kirkpatrick et al., 1983, Aarts and Korst, 1988).”
Sequence Analysis Using Logic Regression (2001, methodology): “Properties of the simulated annealing algorithm depend on Markov chain theory, and thus on the set of operations that can be applied to logic trees [Aarts and Korst, 1989].”
On the Origin of Deep Learning (2017, background): “This corresponds to Zeroth Law of Thermodynamics. algorithm called Simulated Annealing (Khachaturyan et al., 1979; Aarts and Korst, 1988) back then, but Simulated Annealing is hardly relevant to nowadays deep learning society.”
A survey of search methodologies and automated system development for examination timetabling (2009, background): “2 Simulated Annealing Simulated Annealing [1] is motivated by the natural annealing process [2].”
Parameter Control in Evolutionary Algorithms (2007, background): “SA is a generate-and-test search technique based on a physical, rather than a biological analogy [1].”
MONTE CARLO METHODS IN GEOPHYSICAL INVERSE PROBLEMS (2002, methodology): “Simulated annealing remains one of the few approaches specifically designed for global optimization problems that has been successfully applied across several disciplines [Aarts and Korst, 1989].”
Optimal power flow: a bibliographic survey II (2012, background): “Aarts and Korst (1989). In addition, SA is relatively easy to implement and therefore suitable for a wide range of problems.”
Robust surgery loading (2008, methodology): “For λ we use the guideline suggested by Aarts & Korst (1989), the number of neighbor solutions.”
Multistage and coordinated planning of the expansion of transmission systems (2004, background): “Algorithms such as simulated annealing (SA) [15], (GA [11], [12], tabu search (TS) [13], greedy randomized adaptive search procedure (GRASP) [14], and hybrid versions belong to this group of methods.”
Solvent entropy-driven searching for protein modeling examined and tested in simplified models. (2001, methodology): “…method we took an application of the common Monte Carlo (MC) method (Metropolis et al., 1953, Kirkpatrick et al., 1983; © Oxford University Press 329 Aarts and Korst, 1989), implemented in a standard way for this hp model [see Unger and Moult for further details of this standard implementation…”
Noise as a Resource for Computation and Learning in Networks of Spiking Neurons (2014, background): “It builds on and expands earlier related work on Boltzmann machines [25], which did not offer the promise of energy-efficient computation that networks of spiking neurons hold, but which had already pioneered the use of networks of stochastic computing elements for solving CSPs.”
Computational Statistics Handbook with MATLAB (2001, methodology): “[1953], and it is the foundation for other optimization algorithms such as simulated annealing [Kirkpatrick, Gelatt, and Vechi, 1983; Aarts and Korst, 1989].”