scholarly journals Some Remarks on Replicated Simulated Annealing

2021 ◽  
Vol 182 (3) ◽  
Author(s):  
Vicent Gripon ◽  
Matthias Löwe ◽  
Franck Vermet
Keyword(s):  
Neural Networks ◽  
Simulated Annealing ◽  
Statistical Physics ◽  
Simulated Annealing Algorithm ◽  
Real Data ◽  
Data Bases ◽  
Physics Literature ◽  
Annealing Algorithm

AbstractRecently authors have introduced the idea of training discrete weights neural networks using a mix between classical simulated annealing and a replica ansatz known from the statistical physics literature. Among other points, they claim their method is able to find robust configurations. In this paper, we analyze this so called “replicated simulated annealing” algorithm. In particular, we give criteria to guarantee its convergence, and study when it successfully samples from configurations. We also perform experiments using synthetic and real data bases.

Simulation of the Mean-Zero-Up-Crossing Wave Period Using Artificial Neural Networks Trained With Simulated Annealing Algorithm

2006 ◽  
Author(s):  
H. Bazargan ◽  
H. Bahai ◽  
F. Aryana ◽  
S. F. Yasseri
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Simulated Annealing ◽  
Simulated Annealing Algorithm ◽  
North East ◽  
The North ◽  
Wave Heights ◽  
Significant Wave Heights ◽  
Annealing Algorithm ◽  
Artificial Neural

The aim of this work is to simulate the 3-houly mean zero-up-crossing wave periods (Tzs) of the sea-states of a future period for a location in the North East Pacific. Seven multi-layer artificial neural networks (ANNs) were trained with simulated annealing algorithm. The output of each ANN was used for estimating each of the 7 parameters of a new distribution, described in Appendix A, called hepta-parameter spline proposed for the conditional distribution of the Tz given some significant wave heights and mean zero-up-crossing wave periods. After estimating the parameters of the conditional distributions, the Tzs have been forecasted from the hepta-parameter spline distributions corresponding to the Tzs of the period.

scholarly journals An Improved HotSpot Algorithm and Its Application to Sandstorm Data in Inner Mongolia

2020 ◽  
Vol 2020 ◽  
pp. 1-10
Author(s):  
Ren Qing-dao-er-ji ◽  
Rui Pang ◽  
Yue Chang
Keyword(s):  
Simulated Annealing ◽  
Association Rules ◽  
Inner Mongolia ◽  
Simulated Annealing Algorithm ◽  
Solution Space ◽  
Real Data ◽  
Wine Quality ◽  
Minimum Support ◽  
Support Threshold ◽  
Annealing Algorithm

HotSpot is an algorithm that can directly mine association rules from real data. Aiming at the problem that the support threshold in the algorithm cannot be set accurately according to the actual scale of the dataset and needs to be set artificially according to experience, this paper proposes a dynamic optimization algorithm with minimum support threshold setting: S_HotSpot algorithm. The algorithm combines simulated annealing algorithm with HotSpot algorithm and uses the global search ability of simulated annealing algorithm to dynamically optimize the minimum support in the solution space. Finally, the Inner Mongolia sandstorm dataset is used for experiment while the wine quality dataset is used for verification, and the association rules screening indicators are set for the mining results. The results show that S_HotSpot algorithm can not only dynamically optimize the selection of support but also improve the quality of association rules as it is mining reasonable number of rules.

Sign in / Sign up

Export Citation Format

Share Document