PL
 |
EN
Preferencje help
Polish version English version Język
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 5

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  Metody Monte Carlo
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
1
Content available remote Optimized Method based on Lattice Sequences for Multidimensional Integrals in Neural Networks
Todorov Venelin, Dimov Ivan, Fidanova Stefka
Annals of Computer Science and Information Systems
|
2021
|
Vol. 25
243--246
EN
In this work we investigate advanced stochastic methods for solving a specific multidimensional problem related to neural networks. Monte Carlo and quasi-Monte Carlo techniques have been developed over many years in a range of different fields, but have only recently been applied to the problems in neural networks. As well as providing a consistent framework for statistical pattern recognition, the stochastic approach offers a number of practical advantages including a solution to the problem for higher dimensions. For the first time multidimensional integrals up to 100 dimensions related to this area will be discussed in our numerical study.
2
Content available remote A Practical Solution to Handling Randomness and Imperfect Information in Monte Carlo Tree Search
Świechowski Maciej, Tajmajer Tomasz
Annals of Computer Science and Information Systems
|
2021
|
Vol. 25
101--110
EN
This paper provides practical guidelines for developing strong AI agents based on the Monte Carlo Tree Search algorithm in a game with imperfect information and/or randomness. These guidelines are backed up by series of experiments carried out in the very popular game - Hearthstone. Despite the focus on Hearthstone, the paper is written with reusability and universal applications in mind. For MCTS algorithm, we introduced a few novel ideas such as complete elimination of the so-called nature moves, separation of decision and simulation states as well as a multi-layered transposition table. These have helped to create a strong Hearthstone agent.
3
Content available remote Feasibility of computerized adaptive testing evaluated by Monte-Carlo and post-hoc simulations
Štěpánek Lubomír, Martinková Patricia
Annals of Computer Science and Information Systems
|
2020
|
Vol. 21
359--367
EN
Computerized adaptive testing (CAT) is a modern alternative to classical paper and pencil testing. CAT is based on an automated selection of optimal item corresponding to current estimate of test-taker's ability, which is in contrast to fixed predefined items assigned in linear test. Advantages of CAT include lowered test anxiety and shortened test length, increased precision of estimates of test-takers' abilities, and lowered level of item exposure thus better security. Challenges are high technical demands on the whole test work-flow and need of large item banks. In this study, we analyze feasibility and advantages of computerized adaptive testing using a Monte-Carlo simulation and posthoc analysis based on a real linear admission test administrated at a medical college. We compare various settings of the adaptive test in terms of precision of ability estimates and test length. We find out that with adaptive item selection, the test length can be reduced to 40 out of 100 items while keeping the precision of ability estimates within the prescribed range and obtaining ability estimates highly correlated to estimates based on complete linear test (Pearson’s ρ = 0.96). We also demonstrate positive effect of content balancing and item exposure rate control on item composition.
4
Content available remote A Two-Stage Monte Carlo Approach for Optimization of Bimetallic Nanostructures
Mikhov Rossen, Myasnichenko Vladimir, Kirilov Leoneed, Sdobnyakov Nickolay, Matrenin Pavel, Sokolov Denis, Fidanova Stefka
Annals of Computer Science and Information Systems
|
2020
|
Vol. 21
285--288
EN
In this paper we propose a two-stage lattice Monte Carlo approach for optimization of bimetallic nanoalloys: simulated annealing on a larger lattice, followed by simulated diffusion. Both algorithms are fairly similar in structure, but their combination was found to give significantly better solutions than simulated annealing alone. We also discuss how to tune the parameters of the algorithms so that they work together optimally.
5
Content available remote A New Optimized Adaptive Approach for Estimation of the Wigner Kernel
Todorov Venelin, Fidanova Stefka, Dimov Ivan, Poryazov Stoyan
Annals of Computer Science and Information Systems
|
2020
|
Vol. 21
341--344
EN
In this paper we study numerically an optimized Adaptive Monte Carlo algorithm for the Wigner kernel - an important problem in quantum mechanics represented by difficult multidimensional integrals. We will show the advantages of the optimized Adaptive MC algorithm and compare the results with the Adaptive approach from our previous work [4] and other stochastic approaches for computing the Wigner kernel in 3,6,9-dimensional case. The 12-dimensional case will be considered for the first time. A comprehensive study and an analysis of the computational complexity of the optimized Adaptive MC algorithm under consideration has also been presented.
first rewind previous Strona / 1 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.