Maxim Buzdalov

Learn More
This paper describes an optimization problem with one target function to be optimized and several supporting functions that can be used to speed up the optimization process. A method based on reinforcement learning is proposed for choosing a good supporting function during optimization using genetic algorithm. Results of applying this method to a model(More)
Genetic algorithms (GAs) are widely used in multi-objective optimization for solving complex problems. There are two distinct approaches for GA design: generational and steady-state algorithms. Most of the current state-of-the-art GAs are generational, although there is an increasing interest to steady-state algorithms as well. However, for algorithms based(More)
We propose a new algorithm for incremental nondominated sorting of two-dimensional points. The data structure which stores non-dominating layers is based on a tree of Cartesian trees. If there are N points in M layers, the running time for of an insertion is O(M(1 + log(N/M)) + logM log(N/ logM)), which is O(N) in the worst case. This algorithm can be a(More)
Generation of performance tests for programming challenge tasks is considered. A number of evolutionary approaches are compared on two different solutions of an example problem. It is shown that using helper-objectives enhances evolutionary algorithms in the considered case. The general approach involves automated selection of such objectives.
Worst-case execution time tests can be tricky to create for various computer science algorithms. To reduce the amount of human effort, authors suggest using search-based optimization techniques, such as genetic algorithms. This paper addresses difficult test generation for several maximum flow algorithms from the augmenting path family. The presented(More)
In this paper, an automated method for generation of tests against inefficient solutions for programming challenge tasks on graph theory is proposed. The method is based on the use of (1+1) evolution strategy and is able to defeat several kinds of inefficient solutions. The proposed method was applied to a task from the Internet problem archive, the Timus(More)
In this paper further investigation of the previously proposed method of speeding up single-objective evolutionary algorithms is done. The method is based on reinforcement learning which is used to choose auxiliary fitness functions. The requirements for this method are formulated. The compliance of the method with these requirements is illustrated on model(More)
A first step towards analyzing runtime complexity of an evolutionary algorithm adaptively adjusted using reinforcement learning is made. We analyze the previously proposed EA + RL method that enhances single-objective optimization by selecting efficient auxiliary fitness functions. Precisely, Random Mutation Hill Climber adjusted with Q-learning using(More)
The non-dominated sorting algorithm by Jensen, generalized by Fortin et al to handle the cases of equal objective values, has the running time complexity of O(N logK−1 N) in the general case. Here N is the number of points, K is the number of objectives and K is thought to be a constant when N varies. However, the complexity was not proven to be the same in(More)