Hiding in Social Networks

The Internet and social media have fueled enormous interest in social network analysis. New tools continue to be developed and used to analyze our personal connections. This raises privacy concerns that are likely to exacerbate in the future. In this project we investigate whether individuals or groups can actively manage their connections to evade social network analysis tools. By addressing this question, the general public may better protect their privacy, oppressed activist groups may better conceal their existence, and security agencies may better understand how terrorists escape detection.

This project was a main subject of my PhD dissertation. My work was supported by a Polish National Science Centre grant Preludium (grant number 2015/17/N/ST6/03686), where I was the principal investigator.

Collaborators:

• Tomasz Michalak (University of Warsaw, University of Oxford)
• Talal Rahwan (New York University Abu Dhabi, Khalifa University of Science and Technology)
• Michael Wooldridge (University of Oxford)
• Esteban Moro (Massachusetts Institute of Technology, Universidad Carlos III de Madrid)
• Yevgeniy Vorobeychik (Washington University in Saint Louis)
• Kai Zhou (Washington University in Saint Louis)

What you might take a look at:

• our article about hiding private relationships, published in Scientific Reports
• our article about hiding individuals and communities, published in Nature Human Behaviour
• my PhD dissertation presenting a broad view of the problem of hiding in social networks
• our work about constructing a covert network, published at AAMAS 2017

Disinformation Attacks in Networks

Disinformation continues to raise concerns due to its increasing threat to society. Nevertheless, a threat of a disinformation-based attack on networks and critical infrastructure is often overlooked. In this project, we consider an attack in which an adversary attempts to manipulate the behavior of members of the social network by spreading disinformation encouraging them to shift their behaviour. Our findings demonstrate that vulnerabilities in critical infrastructure may arise not only from hardware and software, but also from behavioral manipulation.

Collaborators:

• Talal Rahwan (New York University Abu Dhabi)
• Bedoor AlShebli (New York University Abu Dhabi)
• Jimmy Chih-Hsien Peng (National University of Singapore)
• Gururaghav Raman (National University of Singapore)

What you might take a look at:

• our article about disinformation attacks on traffic networks, published in Scientific Reports
• our article about disinformation attacks on power grids, published in PLOS One

Strategic Network Diffusion

Spreading of ideas in a social network is usually modeled in the literature as a stochastic process. However, in many real-life applications the exact course of the diffusion can be guided by a certain individual or authority. In this project, we analyze a setting in which the process of spreading of information or influence is entirely under the control of a strategic player.

This is a joint project between Masdar Institute and Massachusetts Institute of Technology, principal investigators are Aamena Alshamsi and Cesar Hidalgo. In July 2017 I got involved as a Post-Doctoral Fellow.

Collaborators:

• Aamena Alshamsi (Massachusetts Institute of Technology, Khalifa University of Science and Technology)
• Cesar A. Hidalgo (Massachusetts Institute of Technology)
• Khaled Elbassioni (Khalifa University of Science and Technology)
• Flavio L. Pinheiro (NOVA University Lisbon, Massachusetts Institute of Technology)
• Jarosław Jankowski (West Pomeranian University of Technology)
• Piotr Bródka (West Pomeranian University of Technology, Wrocław University of Science and Technology)
• Radosław Michalski (West Pomeranian University of Technology, Wrocław University of Science and Technology)

What you might take a look at:

• our article about computational aspects of strategic network diffusion, published in Theoretical Computer Science
• our article about defending network from a strategic attacker, published in ACM TIST
• our article about strategic distribution of supporting seeds, published in PLOS One

Dollar Auction with Spiteful Bidders

The dollar auction is an auction model used to analyze the dynamics of conflict escalation. Unlike in most auction formats, in the dollar auction even players who did not get the item may have to pay a certain price. In this project, we analyze the course of an auction when participating players are spiteful, i.e., they are motivated not only by their own profit, but also by the desire to hurt the opponent. Our results give us insight into the possible effects of meanness onto conflict escalation.

Collaborators:

• Long Tran-Thanh (University of Southampton)
• Tomasz Michalak (University of Warsaw, University of Oxford)
• Agata Nieścieruk (Polish-Japanese Academy of Information Technology)
• Talal Rahwan (Khalifa University of Science and Technology)
• Nicholas Jennings (Imperial College London)

What you might take a look at:

• our first work about a dollar auction between oblivious non-spiteful player and a malicious opponent, published at IJCAI 2015
• our second work about playing a series of dollar auctions using multi-armed bandits algorithm, published at AAMAS 2016
• our third work fully characterizing a dollar auction between two players with various levels of spitefulness in a complete information setting, published at AAAI 2017

Simulating Historical Warfare

Numerous mathematical and computer models are used to simulate a situation on a battlefield. However, most of them intend to simulate modern warfare. I created Petro, a new agent-based model of historical warfare (named after hetman Petro Konashevych-Sahaidachny). It can model various tactics of different kinds of military forces and is simple enough to build massive simulations. I used it to simulate the warfare of 17th century Europe and recreate the Battle of Kokenhausen (1601), testing some of the historians' hypotheses about course of the battle.

This was the subject of my Master thesis and my very first dip in the scientific world. So far I published only one work, but I would love to get back to this subject in the future.

What you might take a look at:

• my work about simulating the battle of Kokenhausen (1601), published at IAT 2014