Welcome to my homepage. I am a postdoctoral research fellow in data science at the Department of Statistics at LSE working with Milan Vojnovic. My research interests are in design and analysis of algorithms for massive data mining problems. Before starting my PhD, I had also worked on exact algorithms for hard combinatorial problems.
I am originally from Sofia, Bulgaria. I studied Computer Science and Mathematics at LMU München and I obtained my PhD in Computer Science from IT University of Copenhagen under the supervision of Rasmus Pagh. My PhD thesis is entitled New Sampling and Sketching Techniques for Data Stream Mining. During my PhD studies I interned at Yahoo Labs in Barcelona where I was hosted by Francesco Bonchi. After my PhD, I was a research scientist at NEC Laboratories Europe where I was part of the Network Data Analytics group. You can visit my Linkedin profile for more details on my education and professional experience.
Below is a list with my publications:
Journal publicationsTriangle counting in dynamic graph streamsAlgorithmica 76(1), 2016 (Extends SWAT 2014 paper with the same title) Alexander Golovnev, Konstantin Kutzkov New exact algorithms for the 2-constraint satisfaction problemTheoretical Computer Science 526, 2014 Konstantin Kutzkov An exact exponential time algorithm for counting bipartite cliquesInformation Processing Letters 112(13), 2012 Konstantin Kutzkov New upper bound for the #3-SAT problemInformation Processing Letters 105(1), 2007 Conference proceedingsMoez Draief, Konstantin Kutzkov, Kevin Scaman, Milan Vojnovic
KONG: Kernels for Ordered Neighborhood Graphs32nd Conference on Neural Information Processing Systems (NeurIPS), 2018
Tian Guo, Konstantin Kutzkov, Mohamed Ahmed, Jean-Paul Calbimonte, Karl Aberer
Efficient Distributed Decision Trees for Robust RegressionMachine Learning and Knowledge Discovery in Databases - European Conference (ECML/PKDD), 2016
Mathias Niepert, Mohamed Ahmed, Konstantin Kutzkov
Learning Convolutional Networks for Graphs33rd International Conference on Machine Learning (ICML), 2016 Konstantin Kutzkov, Mohamed Ahmed, Sofia Nikitaki
Weighted similarity estimation in data streams24th ACM International Conference on Information and Knowledge Management (CIKM), 2015 Konstantin Kutzkov, Rasmus Pagh
Triangle counting in dynamic graph streams14th Scandinavian Symposium and Workshops on Algorithm Theory (SWAT), 2014 Konstantin Kutzkov, Rasmus Pagh
Consistent subset sampling14th Scandinavian Symposium and Workshops on Algorithm Theory (SWAT), 2014 David García-Soriano, Konstantin Kutzkov
Triangle counting in streamed graphs via small vertex covers14th SIAM International Conference on Data Mining (SDM), 2014 Konstantin Kutzkov, Albert Bifet, Francesco Bonchi, Aristides Gionis
STRIP: Stream Learning of Influence Probabilities19th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2013 Konstantin Kutzkov
Deterministic algorithms for skewed matrix products30th International Symposium on Theoretical Aspects of Computer Science (STACS), 2013 Konstantin Kutzkov, Rasmus Pagh
On the Streaming Complexity of Computing Local Clustering Coefficients6th ACM conference on Web Search and Data Mining (WSDM), 2013 Andrea Campagna, Konstantin Kutzkov, Rasmus Pagh On Parallelizing Matrix Multiplication by the Column-Row Method15th Meeting on Algorithm Engineering and Experiments (ALENEX), 2013 Konstantin Kutzkov
Improved Counter Based Algorithms for Frequent Pairs Mining in Transactional Data StreamsMachine Learning and Knowledge Discovery in Databases - European Conference (ECML/PKDD), 2012 Andrea Campagna, Konstantin Kutzkov, Rasmus Pagh Frequent Pairs in Data Streams: Exploiting Parallelism and Skew ICDM Workshops 2011 Alexander S. Kulikov, Konstantin Kutzkov New Bounds for MAX-SAT by Clause LearningSecond International Symposium on Computer Science in Russia (CSR), 2007 Submitted papersKonstantin Kutzkov, Mathias Niepert, Mohamed Ahmed
Scalable Regression Tree Learning in Data StreamsNote that the above papers are provided only for personal use for research or educational purposes. The copyrights belong to the respective publishers. |