Arina Buzdalova

My GECCO 2013 experience

GECCO is one of the largest conferences in the field of genetic and evolutionary computation. I have been working on my bachelor’s and then master’s thesis in this field for the last two years, so I was dreaming about attending this conference. I had previously proposed a method of enhancing evolutionary algorithms with reinforcement learning, which was empirically shown to be efficient. So I was very interested to improve my research with theoretical foundations and I got a great opportunity to do this, thanks to ACM-W and SIGEVO. My attendance was also partially supported by Saint-Petersburg National Research University of Information Technologies, Mechanics and Optics.

I have visited a number of theoretical tutorials and theoretical sessions and listened to talks by such brilliant researches as Thomas Jansen, Frank Neumann, Pietro Oliveto, Carsten Witt, Benjamin Doerr and many more… I was really inspired by their example and learned a lot about useful techniques of runtime analysis and popular theoretical problems.

What is more, I am particularly grateful for the mentorship I was provided during the conference. Christine Zarges got acquainted with my first theoretical result and helped me a lot with her advice, which inspired me to continue my research in the theoretical direction. I am very grateful for her support.

Attending the Student Workshop performed by Tea Tusar and Boris Naujoks was a great experience too. I got a lot of useful questions on my presentation and really enjoyed the detailed discussion of my work. I also presented my work at poster session, which also gave me productive feedback. Another exciting event I attended at GECCO was Women in Evolutionary Computation Meeting organised by Una-May O’Reilly, which was a great chance to improve my networking and make friends.

Attending GECCO has had a great influence on me. Now I’m full of ideas about developing my theoretical research. Together with my husband and colleague Maxim Buzdalov I wrote an article about runtime complexity of my method on a specific problem. I am going to present this article at The 7th Workshop on Theory of Randomised Search Heuristics and I will do my best to publish it in a journal.

I would like to thank Adriana Compagnoni for her support. I am very grateful to ACM Women, SIGEVO and University ITMO for making this unforgettable experience possible!