Friday, August 24, 2018

ICE-TCS Theory Day(s) 2018

The Icelandic Centre of Excellence in Theoretical Computer Science held its 14th annual Theory Day today. The programme is here.

Takeshi Tokuyama (Tohoku University, Japan) kicked off the day with a talk entitled Deformation of Determinants and Related Combinatorics. The talk was based on work that Takeshi did in 1986 (as a mathematics postdoc) and published in 1988. He referred to it as his "sleeping beauty" as there has been a peak of interest in that paper only  in the last ten years. Takeshi's "Weyl character formula" was even the subject of a PROMYS research project in the summer of 2013, which led to a paper that generalizes that formula to the Hall-Littlewood polynomials. The talk surveyed many interesting topics and connections with alternating sign matrices, computation of Boltzmann weights, and crystal bases.

The second half of the Theory Day 2018 was devoted to three short presentations by ICE-TCS researchers devoted to research highlights from the centre. Antonis Achilleos set himself the goal of presenting two and a half years worth of research on monitorability, carried out within the framework of the TheoFoMon project, in 15 minutes. Tigran Tonoyan presented joint work with former ICE-TCS postdoc Christian Konrad addressing the question of whether randomness helps in guessing the middle point of an on-line sequence. (Answer: It does.) Last, but not least, Christian Bean gave an overview of an ongoing project whose goal is to automate proofs of results in enumerative combinatorics. Christian presented many examples of published theorems, some of which from 2018, that can be established using their CombSpecSearcher algorithm and that were previously obtained using human ingenuity.
The Theory Day was preceded by a seminar by Tami Tamir (Efi Arazi School of Computer Science, The Interdisciplinary Center) on Thursday, 23 August 2018. In her talk, Tami gave an excellent introduction to the main research questions in  algorithmic game theory, introduced network formation games and discussed her work with Guy Avni and Orna Kupferman on automata formation games, where the goal of each player is given by a regular language rather than by simple reachability objectives. Tami focused on results presented in this paper, which were mostly of a negative nature. She concluded her informative and entertaining presentation by introducing her university, the Efi Arazi School of Computer Science and how they try to tackle the gender issues in computer science.

Overall, this was another good edition of the Theory Day at ICE-TCS. When we founded the centre on the 29th of April 2005, I had no idea that it would see so many installments of this event. I feel fortunate to have taken part in organizing them, to have worked with so many great colleagues in Iceland and to have met many very interesting and inspiring people along the way. Stay tuned for the Theory Day 2019, which will most likely be held in the spring of next year.

Tuesday, August 07, 2018

2018 Women@McKinsey Dissertation Award to GSSI alumna Yllka Velaj

GSSI alumna Yllka Velaj has been selected as co-recipient of the 2018 Women@McKinsey Dissertation Award (, which is given to a female researcher for a master or doctoral thesis on themes related to Advanced Data Analytics or Machine Learning. Yllka receives the award for her PhD thesis "Information spreading with network augmentation" that she completed at the GSSI under the supervision of Pierluigi Crescenzi (University of Florence) and Gianlorenzo D'Angelo (GSSI). The award comes with a job offer by McKinsey & Company, and with financial support to attend the Open Data Science Conference 2018, which will be held in London in the period 19-22 September 2018.

Citing the explanation of her work given by Yllka in the CWI press release:
"Understanding the dynamics that regulate the diffusion of information in complex networks has been one of the main goals in the field of complex network analysis. The problem, which is also known as influence spreading or information diffusion problem, is motivated by many applications in different fields: from marketing to epidemiology, going through the study of adoption of innovations and the analysis of social networks. In my PhD thesis, I designed new efficient algorithms which use a given budget to suggest new links to be added to the network in order to maximize the diffusion of information”.

Yllka is an Italian scientist of Albanian origin. She obtained her BSc in Computer Science and her MSc in Distributed Systems and Computer Networks from the University of L'Aquila. She was a PhD student in Computer Science in the joint PhD GSSI-IMT programme at the Gran Sasso Science Institute. After one year as a postdoctoral researcher at the University of Chieti-Pescara, in May 2018 she joined the Networks & Optimization group at CWI in Amsterdam as a winner of the fellowship programme promoted by ERCIM - the European Research Consortium for Informatics and Mathematics.

Yllka's research interests include combinatorial algorithms, network analysis, algorithm engineering and recently also algorithmic game theory.

Congratulations to Yllka, her supervisors and the CS group at the GSSI as a whole.