Krzysztof Krawiec

I'm a full professor affiliated to Institute of Computing Science, Poznan University of Technology, where I'm leading the Neurosymbolc Systems Group within the Machine Learning Laboratory. My main fields of interest are program synthesis, machine learning and computer vision -- the latter in particular in applications to medical imaging. See Research and Publications for more details, and my profiles: ResearchGate, DBLP, Google Scholar, ResearcherID, ACM, LinkedIn.

The quite complete list of my more recent publications is available here.

Note: I am unable to keep updating this webpage on a regular basis, please do not assume it to be up-to-date.

2024

• Proud to be honored with another ACM SIGEVO Impact Award, granted by the Executive Committee of the ACM Special Interest Group on Genetic and Evolutionary Computation. The award has been established in 2009, and recognizes papers published at the Genetic and Evolutionary Computation Conference (GECCO) 10 years earlier.
• Happy to join my colleagues on the list of 2% of top-ranked Computer Science researchers, according to a long-term study conducted by Stanford University.
• Proud to see my students Kacper Dobek, Zuzanna Buchnajzer, Stanisław Hapke, and Daniel Jankowski being honored with not one, but two awards in this contest, for their bachelor thesis I've been supervising.
• Gave a keynote at the Medical AI Forum in Taipei (December 9-10 2024), organized by the College od Intelligent Computing at the Chang Gung University, one of leading Taiwanese R&D hubs, focusing on advanced AI and machine learning solutions for biomedical applications.

2023

• As of January 2023, I'm heading the Neurosymbolc Systems Group at the newly formed Machine Learning Laboratory at the Institute of Computing Science. The ML lab was primarily formed from the former members of the Laboratory of Intelligent Decision Support Systems. In connection with that, I have stepped down from the position of the Deputy Head for Science of the Institute, where I served for 6 years.
• In July, I lectured the course Machine Learning Meets Program Synthesis at the ML in PL Summer School organized by the Jagiellonian University in Kraków, Poland. Available on YouTube here.

2022

• Together with my co-authors, we've received the ACM SIGEVO Impact Award, awarded by the Executive Committee of the ACM Special Interest Group on Genetic and Evolutionary Computation. The award has been established in 2009, and recognizes papers published at the Genetic and Evolutionary Computation Conference (GECCO) 10 years earlier.
• My PhD student Iwo Błądek defended his thesis Machine Learning and Formal Verification for Acquisition of Knowledge in Heuristic Program Synthesis.

2021

• I'm serving as the General Chair of the ACM Genetic and Evolutionary Computation Conference (GECCO) in Lille, July 10-14, 2021. The conference proceedings are already available!
• We opened a new Masters Course in Artificial Intelligence (in Polish)
• Our Institute of Computing Science and CAMIL are members of Confederation of Laboratories for Artificial Intelligence Research in Europe (CLAIRE).
• We are partners in the TAILOR network, Foundations of Trustworthy AI – Integrating Learning, Optimization and Reasoning.

2019

• The Ph.D. thesis of my student Paweł Liskowski received honourable mention in the Best Ph.D. Dissertation in Artificial Intelligence Contest organized by the Polish Artificial Intelligence Society
• We recently founded CAMIL, Center for Artificial Intelligence and Machine Learning
• "Ain't Nobody Got Time For Coding: Structure-Aware Program Synthesis From Natural Language", by my students Jakub Bednarek, Karol Piaskowski, and me.

2018

• Our paper "Counterexample-Driven Genetic Programming: Stochastic Synthesis of Provably Correct Programs", by Iwo Błądek, Jerry Swan, John H. Drake and me has been accepted and presented at IJCAI'18
• Together with Paweł Liskowski and Wojciech Jaśkowski, we designed a move predictor for the game of Othello based on a deep neural network that defeats all other 1-ply Othello players proposed to date. The paper, accepted for publication in IEEE Transactions on Games, is available here.
• Our study Segmenting Retinal Blood Vessels with Deep Neural Networks is among the 20 most often downloaded papers in IEEE Transactions of Medical Imaging.

2016/2017

• Our paper Counterexample-Driven Genetic Programming, by Iwo Błądek, Jerry Swan and me, won the Best Paper Award at the Genetic and Evolutionary Computation Conference (GECCO'17). The source code we've used to conduct the experiment is available on GitHub.
• According to Springer, the were almost 3000 downloads of the chapters from my book Behavioral Program Synthesis with Genetic Programming (Springer 2016).
• For the third year in a row, together with Malcolm Heywood, we'll be giving the tutorial Solving complex problems with coevolutionary algorithms at GECCO'17.
• My project proposal Software framework for explanatory modeling of big data (LUCID) has been granted funding by The National Centre for Research and Development
• An favorable review of my book Behavioral Program Synthesis with Genetic Programming (Springer 2016) by Atif Azad of University of Limerick, Ireland, has just appeared in Genetic Programming and Evolvable Machines
• As of September 1st, I'm the Deputy Director for Science of the Institute of Computing Science
• Released SwiM (Synthesis with Metaheuristics) - a Scala library for program synthesis using metaheuristics, in particular with genetic programming.
• Our paper Surrogate Fitness via Factorization of Interaction Matrix, by Paweł Liskowski and me, won the Best Paper Award at the 19th European Conference on Genetic Programming (EuroGP)
• The deep neural network approach we designed with my PhD student Paweł Liskowski provides best-to-date segmentation of retinal blood vessel networks in ophthalmological imaging (fundus imaging), outperforming over 70 other methods designed to date. More details in our paper Segmenting Retinal Blood Vessels with Deep Neural Networks, accepted for publication in IEEE Transactions on Medical Imaging. See this page for the online accompanying material.
• My book Behavioral Program Synthesis with Genetic Programming (Springer 2016) is out!
• I'm chairing the Genetic Programming Track at GECCO'16, together with Zdenek Vasicek,
• My daughter Spockified me. Again :)

2015

• Together with Malcolm Heywood, we gave a tutorial Solving complex problems with coevolutionary algorithms at GECCO'15. The slides are available here.
• Together with Alberto Moraglio, we gave a tutorial Semantic Genetic Programming at GECCO'15. The slides are available here.
• The initiative of metaheuristic standardization. More details soon on the Metaheuristics in the Large website.
• This winter, I gave a series of lectures on program synthesis and genetic programming at University of Warsaw. The slides are available here.
• Together with Chris Simons, Jerry Swan, Daniel Tauritz, and Jim Smith, we are organizing MetaDeeP 2015, a workshop on Metaheuristic Design Patterns, collocated with GECCO, Genetic and Evolutionary Computation Conference, in Madrid, Spain, 11-15 July 2015.
• The thesis of my PhD student, Marcin Szubert, entitled "Coevolutionary Shaping for Reinforcement Learning", received first prize in the contest organized by the Polish Artificial Intelligence Society. More details here.
• Together with Alberto Moraglio, we are chairing the Genetic Programming Track at the Genetic and Evolutionary Computation Conference GECCO'15 in Madrid, Spain
• Together with Colin Johnson, Alberto Moraglio, and Michael O'Neill, we are organizing the second edition of the workshop on Semantic Methods in Genetic Programming, this time collocated with the GECCO'15 conference.

2014

• My joint paper with Una-May O'Reilly ''Behavioral Programming: A Broader and More Detailed Take on Semantic GP" won the Best Paper Award in the Genetic Programming track of the GECCO'14 conference.
• From Sept'13 till Feb'14, I was a Fulbright Visiting Scholar at Computer Science and Artificial Intelligence Laboratory (CSAIL) at Massachusetts Institute of Technology (MIT), working in the ALFA group directed by Professor Una-May O'Reilly.
• Together with Jerry Swan, John Clark, Chris Simons, and John Woodward, we are organizing MDP 2014, a workshop on Metaheuristic Design Patterns, collocated with GECCO, Genetic and Evolutionary Computation Conference, in Vancouver, Canada, 12-16 July 2014.
• Together with Colin Johnson, Alberto Moraglio, and Michael O'Neill, we are organizing workshop on Semantic Methods in Genetic Programming at PPSN'14 conference. The workshop will be held as a full-day, four-session event on September 13, 2014, at the GR-Ljubljana Exhibition and Convention Centre
• Together with Miguel Nicolau, we were co-chairing EuroGP 2014, 17th European Conference on Genetic Programming, part of Evo*, Granada, Spain, 23-25 April.

2013 and earlier

• Together with Gisele L. Pappa, John Woodward, and Jerry Swan, we are organizing 3rd Workshop on Evolutionary Computation for the Automated Design of Algorithms at GECCO 2013. Details here.
• The thesis of my former PhD student, Wojciech Jaśkowski, entitled "Algorithms for Test-based Problems", received distinction in the first contest organized by the Polish Artificial Intelligence Society. More details here.
• Our paper entitled "Evolutionary Tuning of Compound Image Analysis Systems for Effective License Plate Recognition", by Mateusz Nawrocki and me, won the best paper award at the ICCCI (The 3rd International Conference on Computational Collective Intelligence - Technologies and Applications),
• Our position evaluation function for the game of Othello (epTDLmpx_12x6) , developed by my student Marcin Szubert, ranks first in the Othello Position Evaluation Function League, organized by Simon Lucas and his team at University of Essex, UK.

Software

• I am developing a evolutionary computation library in Scala Scevo and a genetic programming (program synthesis) library in Scala Scaps
• The MIT Alfa group published the new release of the FlexGP framework, which includes also the implementation of Multiple Regression Genetic Programming, a method we described in this GECCO paper.

Collaborators

• Jerry Swan, University of York, UK
• Una-May O'Reilly, MIT, US
• Alberto Moraglio, University of Exeter, UK
• John Drake, Queen Mary University of London, UK
• Colin Johnson, University of Kent, UK

Postdocs

• Tomasz Pawlak
• Iwo Błądek

Phd students

• Jakub Bednarek,
• Jakub Kwiatkowski
• Piotr Wyrwiński
• Antoni Nowinowski

Former PhD students and postdocs

• Wojciech Jaśkowski
• Marcin Szubert (now at Google)
• Paweł Lichocki (now at Google)
• Mateusz Nawrocki (now at TomTom)
• Bartosz Wieloch,
• Paweł Liskowski