Dr Bart Jaworski

Dr Bart Jaworski

Warszawa, Woj. Mazowieckie, Polska
118 tys. obserwujących 500+ kontaktów

Dołącz, aby obserwować

Informacje

I am a resourceful and experienced Product Manager that, as Mark Twain suggested, never…

Usługi

Poproś o ofertę

Autor artykułów: Dr Bart Jaworski

  • A heartwarming product story
    26 cze 2020

    A heartwarming product story

    I had this in my backlog of posts for a while now, but happy I can finally share it :) I'm currently in a setting…

    3 komentarze
  • Working from home for dummies
    18 mar 2020

    Working from home for dummies

    Hi there! Before working in the office for 4 years now, I have initially worked remotely for 9 years after college…

  • Are job boards dying?
    15 lip 2018

    Are job boards dying?

    Recently I have heard yet another time that "Job boards are dead". It's not even "job boards are in decline" or "We are…

  • I've been a doctor for a year now!
    4 gru 2017

    I've been a doctor for a year now!

    Wow, that went along quickly! I can remember that day like it was yesterday: the great pressure, the audience:…

  • A brand new career step finalized
    26 lis 2017

    A brand new career step finalized

    Now, let's start by saying I'm extremely happy I got here. Leaving a company I have been with since the middle of my…

    2 komentarze
  • A goodbye and a new horizon
    14 sie 2017

    A goodbye and a new horizon

    After nine years of adventures, challenges and tons of personal growth, I have decided it's time to leave my comfort…

    6 komentarzy
See all articles

Wkłady

Dołącz teraz, aby zobaczyć wszystkie reakcje

Aktywność

Dołącz, aby zobaczyć całą aktywność

Doświadczenie

  • The Stepstone Group – grafika

    The Stepstone Group

    Warsaw, Mazowieckie, Poland

  • -

  • -

    Warsaw Metropolitan Area

  • -

    Warsaw

  • -

    Warsaw Area, Poland

  • -

    Warszawa, woj. mazowieckie, Polska

  • -

  • -

  • -

    Gdynia, Pomeranian District, Poland

  • -

Wykształcenie

  • Politechnika Gdańska – grafika

    Politechnika Gdańska

    –

    Aktywność i stowarzyszenia:For my PhD I have developed a path planning evolutionary algorithm for high dense traffic maritime areas.

  • –

  • –

  • –

    Aktywność i stowarzyszenia:Erasmus exchange semester

Licencje i certyfikaty

Publikacje

  • Application of Genetic Algorithms in Graph Searching problems

    2nd International Conference on Information Technology, 28-30 June 2010, Gdańsk, Poland.

    Graph searching is a common approach to solving a problem of capturing a hostile intruder by a group of mobile agents. We assume that this task is performed in an environment which we are able to model as a graph G. The question asked is how many agents are needed to capture an arbitrary fast, invisible and smart intruder. This number is called the (edge) search number of G. The strategy which must be performed by agents is called the (edge) search strategy. Unfortunately calculating both the…

    Graph searching is a common approach to solving a problem of capturing a hostile intruder by a group of mobile agents. We assume that this task is performed in an environment which we are able to model as a graph G. The question asked is how many agents are needed to capture an arbitrary fast, invisible and smart intruder. This number is called the (edge) search number of G. The strategy which must be performed by agents is called the (edge) search strategy. Unfortunately calculating both the optimal search strategy and the search number is NP-hard for general graphs. Furthermore, due to the complexity of the pursuit rules, the application of heuristic solutions is not straightforward. In this paper we suggest a method of applying genetic algorithms to solve graph searching problem. The idea is based on LaPaugh's result on graph searching monotonicity and utilizes representation of a search strategy as a permutation of edges.

    Other authors
    • Łukasz Wrona
    See publication

  • Comparison of reproductive strategies in genetic algorithm approach to graph searching

    2nd International Conference on Information Technology, 28-30 June 2010

    Genetic algorithms (GA) are a well-known tool used to obtain approximate solutions to optimization problems. Successful application of a genetic algorithm in solving given problem is largely dependant on selecting appropriate genetic operators. Selection, mutation and crossover techniques play a fundamental role in both time needed to obtain results and their accuracy. In this paper we focus on applying genetic algorithms in calculating (edge) search number and search strategy for general…

    Genetic algorithms (GA) are a well-known tool used to obtain approximate solutions to optimization problems. Successful application of a genetic algorithm in solving given problem is largely dependant on selecting appropriate genetic operators. Selection, mutation and crossover techniques play a fundamental role in both time needed to obtain results and their accuracy. In this paper we focus on applying genetic algorithms in calculating (edge) search number and search strategy for general graphs. Our genetic representation of problem domain is based on representing search strategy as a permutation of edges and fitness function is based on the number of searchers needed to perform a given strategy. Our implementation of GA is utilized to compute search strategies for selected graph classes. We compare and discuss results obtained while employing different reproduction strategies.

    Other authors
    • Łukasz Wrona
    See publication

  • Comparison of selection schemes in evolutionary method of path planning

    Lecture Notes in Artificial Intelligence : Computational Collective Intelligence : Technologies and Applications. – LNAI 6923, 3rd International Conference on Computational Collective Intelligence - Technologies and Applications, ICCCI 2011

    This article compares an impact of using various selection schemes on the quality of the solution for the problem of planning the path for a moving object using the evolutionary method. In study case problem of avoiding collisions at sea is analyzed. The modelled environment includes static constraints (lands, canals, etc.) and dynamic objects (moving ships). Article analyses behaviour of selection schemes in two similar environments which differ in number of dynamic objects (highly congested…

    This article compares an impact of using various selection schemes on the quality of the solution for the problem of planning the path for a moving object using the evolutionary method. In study case problem of avoiding collisions at sea is analyzed. The modelled environment includes static constraints (lands, canals, etc.) and dynamic objects (moving ships). Article analyses behaviour of selection schemes in two similar environments which differ in number of dynamic objects (highly congested areas). Research has proven that application of specific selectors improves results of study case evolutionary path planning situation.

    Other authors
    • Piotr Kolendo
    • Roman Śmierzchalski
    See publication

  • Evolutionary hierarchical agent decision support system for marine traffic coordination

    9th IFAC Conference on Manoeuvring and Control of Marine Craft 2012

    This paper focuses on presenting the principals and early research on a modification of the Evolutionary Planner/Navigator (EP/N) system, that will introduce an agent system, that based on a hierarchy established with the help of AIS and radar, also incorporating COLREGs rules will be presenting a coordinated traffic plan for all ships in the analysed environment. The idea of this concept will be presented along with its requirements and research already undertaken. The new ahEP/N will maintain…

    This paper focuses on presenting the principals and early research on a modification of the Evolutionary Planner/Navigator (EP/N) system, that will introduce an agent system, that based on a hierarchy established with the help of AIS and radar, also incorporating COLREGs rules will be presenting a coordinated traffic plan for all ships in the analysed environment. The idea of this concept will be presented along with its requirements and research already undertaken. The new ahEP/N will maintain all of the benefits of the original algorithm (such as near real time work) and will allow the coordination of safe and economic path planning between all the ships in the area overseen by the watchkeeper.

    Other authors
    • Roman ŚmierzchalskiŁ
    • Łukasz Kolendo
    See publication

  • Experimental research on evolutionary path planning algorithm with fitness function scaling for collision scenarios

    Gdynia Maritime University, TransNav 2011, Gdynia

    This article presents typical ship collision scenarios, simulated using the evolutionary path planning system and analyses the impact of the fitness function scaling on the quality of the solution. The function scaling decreases the selective pressure, which facilitates leaving the local optimum in the calcula-tion process and further exploration of the solution space. The performed investigations have proved that the use of scaling in the evolutionary path planning method makes it possible to…

    This article presents typical ship collision scenarios, simulated using the evolutionary path planning system and analyses the impact of the fitness function scaling on the quality of the solution. The function scaling decreases the selective pressure, which facilitates leaving the local optimum in the calcula-tion process and further exploration of the solution space. The performed investigations have proved that the use of scaling in the evolutionary path planning method makes it possible to preserve the diversity of solu-tions by a larger number of generations in the exploration phase, what could result in finding better solution at the end. The problem of avoiding collisions well fitted the algorithm in question, as it easily incorporates dy-namic objects (moving ships) into its simulations, however the use scaling with this particular problem has proven to be redundant.

    Side note: This paper was granted "Best paper award" on the Transnav 2011 conference.

    Other authors
    • Łukasz Kolendo
    • Roman Śmierzchalski
    See publication

  • Extinction Event Concepts for the Evolutionary Algorithms

    Extinction Event Concepts for the Evolutionary Algorithms

    he main goal of this present paper is to propose a structure for a tool helping to determine how algorithm would react in a real live application, by checking it’s adaptive capabilities in an extreme situation. Also a different idea of an additional genetic operator is being presented. As Genetic Algorithms are directly inspired by evolution, extinction events, which are elementary in our planet’s development history, became a foundation for those concepts.

    Other authors
    • Śmierzchalski Roman
    • Łukasz Kolendo
    See publication

  • Fitness function scaling in the evolutionary method of path planning

    21st IEEE International Symposium on Industrial Electronics

    The article analyses the impact of fitness function scaling on the quality of the solution to the problem of planning the path for a moving object using the evolutionary method. The function scaling decreases the selective pressure, which facilitates leaving the local optimum in the calculation process and further exploration of the solution space. The performed investigations have proved that the use of scaling in the evolutionary path planning method makes it possible to preserve the…

    The article analyses the impact of fitness function scaling on the quality of the solution to the problem of planning the path for a moving object using the evolutionary method. The function scaling decreases the selective pressure, which facilitates leaving the local optimum in the calculation process and further exploration of the solution space. The performed investigations have proved that the use of scaling in the evolutionary path planning method makes it possible to preserve the diversity of solutions by a larger number of generations in the exploration phase. The problem of avoiding collisions at sea was selected as the test case. The modelled environment includes static constraints (lands, canals, etc.) and dynamic objects (moving ships).

    Other authors
    • Roman Śmierzchalski
    • Łukasz Kolendo
    See publication

  • Mean crossover in evolutionary path planning method for maritime collision avoidance

    ExploShip 2012, Świnoujście

    This paper presents the use of mean crossover genetic operator for path planning using evolutionary algorithm for collision avoidance on sea. Mean crossover ensures widening of the possible solutions’ set that can be achieved in comparison to exchange crossover variant. The research shown, that the mean crossover allows to achieve results independent from the initial generation and quicker transition of the algorithm from the exploration to the exploitation phase. New version of the algorithm…

    This paper presents the use of mean crossover genetic operator for path planning using evolutionary algorithm for collision avoidance on sea. Mean crossover ensures widening of the possible solutions’ set that can be achieved in comparison to exchange crossover variant. The research shown, that the mean crossover allows to achieve results independent from the initial generation and quicker transition of the algorithm from the exploration to the exploitation phase. New version of the algorithm allows for an effective solution search for the problem of a collision scenario on sea.

    Other authors
    • Łukasz Kolendo
    • Roman Śmierzchalski
    • Kuczkowski Łukasz
    See publication

  • Methods of selective pressure control in evolutionary path planning

    16th International Conference on Methods and Models in Automation and Robotics, s. 413 - 420, MMAR 2011, Międzyzdroje

    Other authors
    • Piore Kolendo
    • Roman Śmierzchalski

  • Skalowanie funkcji przystosowania w ewolucyjnej metodzie planowania ścieżek przejścia

    Wydaw. Politechniki Świętokrzystkiej, 2011. - (Monografie, t. 16). - S. 66-7

    Other authors
    • Piotr Kolendo
    • Roman Śmierzchalski

  • Zastosowanie krzyżowania uśredniającego do ewolucyjnej metody wyznaczania ścieżki przejścia na morzu

    Oddział Warszawski Polskiego Towarzystwa Elektrotechniki Teoretycznej i Stosowanej, PTETIS 2012, Kościelisko

    This paper presents the use of mean crossover genetic operator for path i planning using evolutionary algorithm for collision avoidance on sea. ! Mean crossover ensures widening of the possible solutions' set that can be achieved in comparison to exchange crossover variant. The research i shown, that the mean crossover allows to achieve results independent i from the initial generation and quicker transition of the algorithm from the exploration to the exploitation phase. New version of the…

    This paper presents the use of mean crossover genetic operator for path i planning using evolutionary algorithm for collision avoidance on sea. ! Mean crossover ensures widening of the possible solutions' set that can be achieved in comparison to exchange crossover variant. The research i shown, that the mean crossover allows to achieve results independent i from the initial generation and quicker transition of the algorithm from the exploration to the exploitation phase. New version of the algorithm i allows for an effective solution search for the problem of a collision scenario on sea.

    Other authors
    • Piotr Kolendo
    • Śmierzchalski Roman
    • Kuczkowski Łukasz
    See publication

Projekty

  • Great Product Manager - Practical Product Management course

    – obecnie

    Save time with this practical Product Management course! With this program, one can land their dream job/promotion way quicker!

    Throughout over 11 hours of core material, I am not only teach about the position itself but also, I am going to provide you with practical solutions on how to deal with different setbacks that I had to find out the hard way. I am also adding a bonus class every week, starting from 18/04/2021.

    After completing this course, you will be well prepared…

    Save time with this practical Product Management course! With this program, one can land their dream job/promotion way quicker!

    Throughout over 11 hours of core material, I am not only teach about the position itself but also, I am going to provide you with practical solutions on how to deal with different setbacks that I had to find out the hard way. I am also adding a bonus class every week, starting from 18/04/2021.

    After completing this course, you will be well prepared to:

    manage any IT product,
    - choose the best improvements for it, and make them a reality with your team,
    - envision the product's growth for many years to come,
    - build a great product road map,
    - change your stale development team into a well-oiled, highly effective machine,
    - find great ideas for your product and support them with the right data,
    - excel in planning an A/B test for any product change and be able to understand its effects,
    - become a great spokesman,
    - conduct engaging meetings and make them highly effective,
    - complete any product management certification. This course provides a certificate as well upon completion!
    - and many more!

    I have almost ten years of background in working on various IT tools designed for recruiters. Apart from that, I have transitioned from a rookie, self-taught product manager in a small start-up to a senior product manager in, arguably, one of the greatest software companies of our times.

    Thus, this course is highly dedicated to seeing you succeed in your career as a product manager the way I have. I will take you through the whole process of planning your career, starting with building an ideal CV and a LinkedIn profile. I will guide you on how to perform at your best on your potential job interview and start a new product manager position with a BANG.

    See project

  • UPP - Universal Partner Page for idibu

    This is a great integration tool that allows any partner to integrate with idibu with very little effort. It led idibu to sign up many new clients and partners over the years, making this the most successful product I have ever designed and seen through. This was designed as an ultimate solution to all integration challenges idibu ever faced. It begun as a side project and quickly grew to one of the most important parts of idibu's product portfolio. Despite low development effort, this project…

    This is a great integration tool that allows any partner to integrate with idibu with very little effort. It led idibu to sign up many new clients and partners over the years, making this the most successful product I have ever designed and seen through. This was designed as an ultimate solution to all integration challenges idibu ever faced. It begun as a side project and quickly grew to one of the most important parts of idibu's product portfolio. Despite low development effort, this project addressed the following problems that jeopardised past integrations:

    - Little to no development and post-go-live support necessity: Tool is so easy to work with, that external partners had little problems understanding the technical requirements. Maintaining it was also not troublesome.

    - Integration speed: The way the integration tool was built, cut down the integration time from months and weeks to days and, for the record holders, hours. That meant low integration costs on partners' end.

    - Optimal UX: Using UPP forced partners to provide an optimal multi-posting interface, thus reducing the training effort on the partners' end increasing the quality of the posted ads.

    - Focus on the user: This integration model provides a great separation on what is required from a functional integration and config.

    - Ease of styling with CSS.

    - Compatibility with idibu's different sales models.

    The inception of this project was preceded by different integration models and experiences, that lead me to provide the design for this ultimate solution. The success of this model also carried on to the newest idibu version (link: https://github.com/oneworldmarket/idibu-v3-api/tree/master/stuff/iFrame%20integration) which I helped on before I left the company. The final project I worked on with the UPP of the new idibu was the very successful idibu-bullhorn integration (link: https://www.youtube.com/watch?v=XFa9LFrKybo) which is a part of Bullhorn's marketplace.

    Other creators
    See project

  • Hierarchical evolutionary path planning for maritime objects in high density traffic areas

    –

    For my doctoral project, input data such as current position, course, and speed of ships that move in the area under VTS (Vessel Traffic System) surveillance can be obtained via AIS (Automatic Identification System). This data allows one to detect possible collision scenarios and calculate a set of safe paths for all the observed objects. COLREGs rules are also considered. Those rules determine which of the 2 ships in a potential collision scenario gives way to the other vessel. Therefore one…

    For my doctoral project, input data such as current position, course, and speed of ships that move in the area under VTS (Vessel Traffic System) surveillance can be obtained via AIS (Automatic Identification System). This data allows one to detect possible collision scenarios and calculate a set of safe paths for all the observed objects. COLREGs rules are also considered. Those rules determine which of the 2 ships in a potential collision scenario gives way to the other vessel. Therefore one can determine a hierarchy of any number of ships. Thus, those ships can be described as an ordered set of vessels in the given environment. Ships' position in this set determines which vessel is giving way according to COLREGs rules.
    Those principles were adopted in the hEP/N algorithm, which is a computer program that applies the proposed method.
    Hierarchical evolutionary path planning method consists of the following steps :

    1. Acquisition of the input data describing the vessels in the observed environment to the hEP/N algorithm.
    2. Calculation of the ships’ hierarchy using COLREGs rules.
    3. Hierarchy acceptance. If the calculated hierarchy does not match the current navigational scenario, it is possible for the watchkeeper to modify it.
    4. Path planning for all the ships in the observed environment.
    5. Path implementation

    Path planning is executed using an evolutionary algorithm (link: https://en.wikipedia.org/wiki/Evolutionary_algorithm).

    Hierarchical evolutionary path planning, in comparison to existing ships' path planning algorithms, is an improvement due to the use of the hierarchy mechanism. It allows easy and reliable implementation of the COLREGs rules regardless of the number of ships in the observed environment. There is another way this method is unique due to hierarchy: The hierarchy can be manually edited so that the method can provide optimal paths when the basic COLREGs rules can't apply.

    Other creators
    • Roman Śmierzchalski
    See project

Wyróżnienia i nagrody

  • Best Paper Award

    TransNav 2011

    My paper "Experimental research on evolutionary path planning algorithm with fitness function scaling for collision scenarios" was awarded for excellent paper and presentation quality.

Języki

  • English

    Biegłość na poziomie zaawansowanym

  • Polish

    Język ojczysty lub biegłość dwujęzyczna

Otrzymane rekomendacje

30 osób poleciło użytkownika Dr Bart Jaworski

Dołącz teraz, aby wyświetlić

Więcej działań użytkownika Dr Bart Jaworski

Wyświetl pełny profil użytkownika Dr Bart Jaworski.

  • Zobacz, jakich macie wspólnych znajomych
  • Zostań przedstawiony(-a)
  • Skontaktuj się bezpośrednio z użytkownikiem Dr Bart Jaworski
Dołącz, aby wyświetlić pełny profil

Inne podobne profile

Odkryj więcej publikacji

Dodaj nowe umiejętności dzięki tym kursom

Zobacz wszystkie kursy