Jiri Kubalik

Jiri Kubalik
Czech Technical University in Prague | ČVUT · Czech Institute of Informatics, Robotics and Cybernetics

PhD

About

79
Publications
6,748
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
482
Citations
Additional affiliations
October 1996 - present
Czech Technical University in Prague
Position
  • Professor (Assistant)

Publications

Publications (79)
Article
Full-text available
The facility layout problem is one of the fundamental production system management problems. It has a significant impact on overall system efficiency. This paper introduces a new facility layout problem that allows for choosing from multiple variants of each facility. The need for choosing the most suitable selection from the facility variants whil...
Preprint
Many real-world systems can be described by mathematical formulas that are human-comprehensible, easy to analyze and can be helpful in explaining the system's behaviour. Symbolic regression is a method that generates nonlinear models from data in the form of analytic expressions. Historically, symbolic regression has been predominantly realized usi...
Article
Full-text available
Many real-world systems can be described by mathematical models that are human-comprehensible, easy to analyze and can be helpful in explaining the system’s behavior. Symbolic regression is a method that can automatically generate such models from data. Historically, symbolic regression has been predominantly realized by genetic programming, a meth...
Preprint
Full-text available
Novel view synthesis is a long-standing problem. In this work, we consider a variant of the problem where we are given only a few context views sparsely covering a scene or an object. The goal is to predict novel viewpoints in the scene, which requires learning priors. The current state of the art is based on Neural Radiance Fields (NeRFs), and whi...
Article
Virtually all dynamic system control methods benefit from the availability of an accurate mathemjThis includes also methods like reinforcement learning, which can be vastly sped up and made safer by using a dynamic system model. However, obtaining a sufficient amount of informative data for constructing dynamic models can be difficult. Consequently...
Article
Full-text available
Continual model learning for nonlinear dynamic systems, such as autonomous robots, presents several challenges. First, it tends to be computationally expensive as the amount of data collected by the robot quickly grows in time. Second, the model accuracy is impaired when data from repetitive motions prevail in the training set and outweigh scarcer...
Article
Full-text available
Reinforcement learning algorithms can solve dynamic decision-making and optimal control problems. With continuous-valued state and input variables, reinforcement learning algorithms must rely on function approximators to represent the value function and policy mappings. Commonly used numerical approximators, such as neural networks or basis functio...
Preprint
Full-text available
This paper deals with the problem of autonomous navigation of a mobile robot in an unknown 2D environment to fully explore the environment as efficiently as possible. We assume a terrestrial mobile robot equipped with a ranging sensor with a limited range and 360 degrees field of view. The key part of the exploration process is formulated as the d-...
Article
Developing mathematical models of dynamic systems is central to many disciplines of engineering and science. Models facilitate simulations, analysis of the system’s behavior, decision making and design of automatic control algorithms. Even inherently model-free control techniques such as reinforcement learning (RL) have been shown to benefit from t...
Preprint
In symbolic regression, the search for analytic models is typically driven purely by the prediction error observed on the training data samples. However, when the data samples do not sufficiently cover the input space, the prediction error does not provide sufficient guidance toward desired models. Standard symbolic regression techniques then yield...
Chapter
Facility layout problems, i.e., optimal placement of production units in a plant, become an inseparable part of manufacturing systems design and management. They are known to greatly impact the system performance. This paper proposes a new formulation of the facility layout problem where workstations are to be placed into a hall. Within the hall, o...
Preprint
Full text of the preprint available at https://arxiv.org/abs/1903.11483 | Reinforcement learning (RL) is a widely used approach for controlling systems with unknown or time-varying dynamics. Even though RL does not require a model of the system, it is known to be faster and safer when using models learned online. We propose to employ symbolic regre...
Preprint
Full text of the preprint available at https://arxiv.org/abs/1903.09688 | Reinforcement learning algorithms can be used to optimally solve dynamic decision-making and control problems. With continuous-valued state and input variables, reinforcement learning algorithms must rely on function approximators to represent the value function and policy ma...
Article
Full-text available
This paper deals with the problem of autonomous navigation of a mobile robot in an unknown2D environment to fully explore the environment as efficiently as possible. We assume a terrestrial mobilerobot equipped with a ranging sensor with a limited range and 360º field of view. The key part of theexploration process is formulated as the d-Watchman R...
Article
Approximate Reinforcement Learning (RL) is a method to solve sequential decisionmaking and dynamic control problems in an optimal way. This paper addresses RL for continuous state spaces which derive the control policy by using an approximate value function (V-function). The standard approach to derive a policy through the V-function is analogous t...
Article
This paper addresses the problem of deriving a policy from the value function in the context of critic-only reinforcement learning (RL) in continuous state and action spaces. With continuous-valued states, RL algorithms have to rely on a numerical approximator to represent the value function. Numerical approximation due to its nature virtually alwa...
Conference Paper
Genetic programming (GP) is a technique widely used in a range of symbolic regression problems, in particular when there is no prior knowledge about the symbolic function sought. In this paper, we present a GP extension introducing a new concept of local transformed variables, based on a locally applied affine transformation of the original variabl...
Article
Model-based reinforcement learning (RL) algorithms can be used to derive optimal control laws for nonlinear dynamic systems. With continuous-valued state and input variables, RL algorithms have to rely on function approximators to represent the value function and policy mappings. This paper addresses the problem of finding a smooth policy based on...
Article
State-of-the-art critic-only reinforcement learning methods can deal with a small discrete action space. The most common approach to real-world problems with continuous actions is to discretize the action space. In this paper a method is proposed to derive a continuous-action policy based on a value function that has been computed for discrete acti...
Chapter
This paper presents a first step of our research on designing an effective and efficient GP-based method for symbolic regression. First, we propose three extensions of the standard Single Node GP, namely (1) a selection strategy for choosing nodes to be mutated based on depth and performance of the nodes, (2) operators for placing a compact version...
Conference Paper
Full-text available
Generally, we distinguish between two classes of hyper-heuristic approaches, heuristic selection and heuristic generation. The former one works with existing heuristics and tries to find their optimal order for solving the instance. The later approach automatically generates new heuristic. Here, these two approaches are combined so that, first, a n...
Conference Paper
The main motivation of this paper is a support of knowledge management for small to medium enterprises (business). We present our tool sitIT.cz which was developed to support communication of IT specialists (both from academia and business) using public funding. The main message of this paper is that this tool is quite generic and can be used in di...
Article
Full-text available
This paper presents a novel evolutionary algorithm for solving routing and sequencing problems. It builds directly on the observation that the optimal solution is composed mostly of short and low-cost links. It uses an indirect representation and an extended nearest neighbor constructive procedure. The representation scheme is redundant making it p...
Conference Paper
Full-text available
In this paper a novel constructive hyperheuristic for CVRP is proposed. This hyperheuristic, called HyperPOEMS, is based on an evolutionary-based iterative local search algorithm. Its inherent characteristics make it capable of autonomously searching a structured space of low-level domain specific heuristics for their suitable combinations that pro...
Conference Paper
Manual design of motion patterns for legged robots is difficult task often with suboptimal results. To automate this process variety of approaches have been tried including various evolutionary algorithms. In this work we present an algorithm capable of generating viable motion patterns for multi-legged robots. This algorithm consists of two evolut...
Article
This paper describes the concept and some preliminary experiments of extension of the sitIT.cz portal-the social network of the ICT specialists in Czech Republic. SitIT.cz interconnects ICT specialists and offers effective search according to several types of structured-machine readable-profiles. It is intended to support technology transfer, shari...
Article
Recently, a new co-evolutionary approach for generating motion patterns for multi-legged robots which exhibit symmetry and module repetition was proposed. The algorithm consists of two evolutionary algorithms working in co-evolution. The first one, a genetic programming module, evolves a motion of a single leg. The second one, a genetic algorithm m...
Chapter
In this paper, we describe the current state of the development of a web portal SitIT.cz. The portal is being developed in the scope of a EU-funded regional project SOSIREČR ( http://www.sosirecr.cz ). It is based on the concept of a social network which has become a very common concept in recent years. It differs from the existing portals in its s...
Article
Abstract Six population-based methods for real-valued black box optimization are thoroughly compared in this article. One of them, Nelder-Mead simplex search, is rather old, but still a popular technique of direct search. The remaining five (POEMS, G3PCX, Cauchy EDA, BIPOP-CMA-ES, and CMA-ES) are more recent and came from the evolutionary computati...
Conference Paper
Full-text available
This paper proposes an evolutionary-based iterative local search hyper-heuristic approach called Iterated Search Driven by Evolutionary Algorithm Hyper-Heuristic (ISEA). Two versions of this algorithm, ISEA-chesc and ISEA-adaptive, that differ in the re-initialization scheme are presented. The performance of the two algorithms was experimentally ev...
Conference Paper
This paper describes the concept of a social network of the ICT specialists in the regions of the Czech Republic. In particular, we focus on the web portal under development, i.e. a software tool serving for the network implementation. Associated activities concerning collecting and analyzing ICT requirements from companies and educational ICT know...
Conference Paper
Full-text available
The Shortest Common Supersequence (SCS) problem is a well-known hard combinatorial optimization problem with applications in many areas. This paper presents two extensions of recently proposed evolutionary-based iterative local search algorithm called POEMS for solving the SCS problem. Both extensions improve scalability of the algorithm. The first...
Article
Large software companies have to plan their project portfolio to maximize potential portfolio return and strategic alignment, while balancing various preferences, and considering limited resources. Project portfolio managers need methods and tools to find a good solution for complex project portfolios and multiobjective target criteria efficiently....
Conference Paper
We describe a new approach to the application of stochastic search in Inductive Logic Programming (ILP). Unlike traditional approaches we do not focus directly on evolving logical concepts but our refinement-based approach uses the stochastic optimization process to iteratively adapt the initial working concept. Utilization of context-sensitive co...
Conference Paper
Full-text available
The Shortest Common Supersequence (SCS) problem is a well-known hard combinatorial optimization problem that formalizes many real world problems. Recently, an application of the iterative optimization method called Prototype Optimization with Evolved Improvement Steps (POEMS) to the SCS problem has been proposed. The POEMS seeks the best variation...
Conference Paper
Full-text available
Large software companies have to plan their project portfolio to maximize potential portfolio return and strategic alignment, while balancing various preferences, and considering limited resources. However, software project portfolios are challenging to describe for optimization in a practical way that allows efficient optimization. In this paper w...
Conference Paper
Full-text available
The paper presents a successful application of an evolutionary based iterative optimization method called Prototype Optimization with Evolved Improvement Steps (POEMS) to the DNA fragment assembly problem. The DNA fragment assembly problem, known to be NP-hard, is of great importance as it constitutes an important step in the genome project. The PO...
Conference Paper
Full-text available
Estimation-of-distribution algorithm using Cauchy sampling distribution is compared with the iterative prototype optimization algorithm with evolved improvement steps. While Cauchy EDA is better on unimodal functions, iterative prototype optimization is more suitable for multimodal functions. This paper compares the results for both algorithms in m...
Article
This paper presents benchmarking of a stochastic local search algorithm called Prototype Optimization with Evolved Improvement Steps (POEMS) on the BBOB 2010 noise-free functions testbed. An original version of the POEMS algorithm presented at BBOB 2009 workshop is compared to a new variant using a pool of candidate prototypes. Experiments for 2D,...
Article
Multiagent systems consist of a collection of agents that directly interact usually via a form of message passing. Information about these interactions can be analyzed in an online or offline way to identify clusters of agents that are related. The first part of this paper is dedicated to a formal definition of a proposed dynamic model for agent cl...
Conference Paper
Full-text available
This paper presents an application of a prototype optimization with evolved improvement steps algorithm (POEMS) to the well-known problem of optimal sorting network design. The POEMS is an iterative algorithm that seeks the best variation of the current solution in each iteration. The variations, also called hypermutations, are evolved by means of...
Conference Paper
Full-text available
This paper presents benchmarking of a stochastic local search algorithm called Prototype Optimization with Evolved Improvement Steps (POEMS) on the noise-free BBOB 2009 testbed. Experiments for 2, 3, 5, 10 and 20 D were done, where D denotes the search space dimension. The maximum number of function evaluations is chosen as 105 x D. Experimental re...
Conference Paper
Full-text available
This paper deals with a Multiple Sequence Alignment problem, for which an implementation of the Prototype Optimization with Evolved Improvement Steps (POEMS) algorithm has been proposed. The key feature of the POEMS is that it takes some initial solution, which is then iteratively improved by means of what we call evolved hypermutations. In this wo...
Conference Paper
This paper proposes a novel grammar-based framework of concept representation for randomized search in relational learning (RL), namely for inductive logic programming. The utilization of grammars guarantees that the search operations produce syntactically correct concepts and that the background knowledge encoded in the grammar can be used both fo...
Article
Grammatical Evolution (GE) is an evolutionary-based optimization technique that evolves tree-like program structures in an arbitrary language encoded as linear chromosomes. GE adopts a genotype-phenotype mapping process taking as input a grammar that describes the syntax of the evolved programs. An advantage of GE is that for searching the solution...
Conference Paper
Full-text available
Recently, a new iterative optimization framework utilizing an evolutionary algorithm called "Prototype Optimization with Evolved iMprovement Steps" (POEMS) was introduced, which showed good per- formance on hard optimization problems - large instances of TSP and real-valued optimization problems. Especially, on discrete optimization problems such a...
Conference Paper
Full-text available
This paper presents a new approach for solving network flow optimization problems. In particular, the goal is to op- timize the traffic in the network structured event-driven sys- tems as well as to provide means for efficient adaptation of the system to changes in the environment - i.e. when some nodes and/or links fail. Many network flow optimiza...
Chapter
Full-text available
A good performance of traditional genetic algorithm is determined by its ability to identify building blocks and grow them to larger ones. To attain this objective a properly arranged chromosome is needed to ensure that building blocks will survive the application of recombination operators. The proposed algorithm periodically rearranges the order...
Conference Paper
Full-text available
Evolutionary algorithms have already been more or less successfully applied to a wide range of optimisation problems. Typically, they are used to evolve a population of complete candidate solutions to a given problem, which can be further refined by some problem-specific heuristic algorithm. In this paper, we introduce a new framework called Iterat...
Chapter
Full-text available
This paper presents an application of soft computing techniques to the construction of decision support tool used for identifying the economically unstable licensed subjects. The work has been initiated by the Czech Energy Regulatory Office whose main mission is to guard the regular heat supply without significant disturbances. Thus the main goal i...
Conference Paper
Full-text available
A good performance of traditional genetic algorithm is determined by its ability to identify building blocks and grow them to larger ones. To attain this objective a properly arranged chromosome is needed to ensure that building blocks will survive the application of recombination operators. The proposed algorithm periodically rearranges the order...
Conference Paper
Full-text available
Evolutionary algorithms are typically used to evolve a population of complete candidate solutions to a given problem. Recently, a novel framework called iterative prototype optimization with evolved improvement steps has been proposed. This is a general optimization framework, where a possible improvement of a prototype solution is being evolved by...
Article
In medical systems it is often advantageous to utilize specific problem situations (cases) in addition to or instead of a general model. Decisions are then based on relevant past cases retrieved from a case memory. The reliability of such decisions depends directly on the ability to identify cases of practical relevance to the current situation. Th...
Conference Paper
Full-text available
This paper presents an algorithm for induction of ensembles of decision trees, also referred to as decision forests. In order to achieve high expressiveness the trees induced are multivariate, with various, possibly user-defined tests in their internal nodes. Strongly typed genetic programming is utilized to evolve structure of the tests. Special a...
Chapter
Full-text available
This paper presents genetic algorithms with real-coded binary representation - a novel approach to improve the performance of genetic algorithms. The algorithm is capable of maintaining the diversity of the evolved population during the whole run which protects it from the premature convergence. This is achieved by using a special encoding scheme,...
Conference Paper
This paper presents an application of ant colony optimisation and genetic algorithm to rescue operation planning. It considers the task as the multiple travelling salesmen problem and proposes suitable heuristics in order to improve the performance of the selected techniques. Then it applies the implemented solutions to a real data. The paper concl...
Article
Full-text available
Production process control becomes complicated as the complexity of the controlled process grows. To simplify the operators role many computer based control systems with integrated visualization clients have been developed. In many practical circumstances malfunction of one or more process components results in other related components entering the...
Conference Paper
Full-text available
Grammatical evolution is an evolutionary algorithm designed to evolve programs in any language. Grammatical evolution operates on binary strings and the mapping of the genotype onto the phenotype (the tree representation of the programs) is provided through the grammar described in the form of production rules. The program trees are constructed in...
Conference Paper
Full-text available
For patients considering elective major surgery, information about operative mortality risks is essential for careful decision making. To help patients and surgeons make informed decisions about whether to undergo elective high-risk surgery, a reliable predictive model would be beneficial. This paper focuses on development and optimized tuning of a...
Conference Paper
Full-text available
This paper presents an application of genetic algorithm to a problem of finding of an optimal layout of objects on a material strip with the aim to reduce the total length of material used. In order to reduce the total search space the problem was solved as an ordering problem. This paper focuses on the representation issues of the problem and on d...
Article
Full-text available
This paper describes a genetic programming approach to the construction of fuzzy classification system with if-then fuzzy rules. Recently many research studies were focusing on utilisation of evolutionary techniques for automatically extracting fuzzy rules from data. In this paper we present a method based on genetic programming with a special stru...
Article
Full-text available
This paper discusses a genetic implementation of the growing hyperspheres classifier (GHS) for high- dimensional data classification. The main idea of the GHS classifier consists in data separation by n-dimensional hyperspheres properly spread over the training data. First, the idea of training data representation is described. Then a brief descrip...
Article
Full-text available
The paper describes an enhancement of the traditional 2-point crossover operator used for binary representation in genetic algorithms. This operator preserves a schema common to both parent chromosomes. The enhancement of its functionality is in a modified treatment of this common schema. The offspring produced by the modified operator is partially...
Article
Full-text available
Genetic Algorithms (GAs) are adaptive search methods, which try to incorporate the principle of surviving known from nature. They proved to be an efficient instrument for solving many hard problems in different areas, in which the majority of other techniques failed as being weak or not applicable. On the other hand, GAs fight with a number of prob...
Article
Full-text available
This paper deals with static agent cluster-ing analysis within multi-agent systems. The goal is to find clusters of agents within the multi-agent system that minimize communi-cation among different clusters. Originally, a modified fuzzy clustering algorithm was used for this purpose. In this paper, the prob-lem was transformed to multi-objective op...

Network

Cited By