Avast Team, Jun 18, 2020

Viliam Lisy is a senior research scientist in the Technology and Innovation Office at Avast. He also works as an assistant professor at the Artificial Intelligence Center at the Department of Computer Science at the Czech Technical University (CTU) in Prague, where he and his colleagues lead a research group dealing with game theory.

Viliam will be presenting at CyberSec&AI Connected and his talk is entitled ‘Playing Poker with Cybercriminals’. We caught up with him to find out more about his work and what attendees can expect to hear from his presentation. 

We are very much looking forward to welcoming you as a speaker to CyberSec&AI Connected in October. Could you start by giving us an overview of the work you do at the Center for Artificial Intelligence at the Department of Computer Science at the Czech Technical University (CTU) in Prague?

Besides teaching a few courses, I lead a research group focusing on computational game theory. We are developing algorithms for finding near-optimal strategies in situations where multiple agents optimize conflicting goals in a shared environment. The algorithms are general and can be applied to any environment. Therefore, we always test them on multiple domains at the same time, be it card games, board games, or models of real world situations from networks or physical security systems. 

And what kind of areas are you exploring at Avast, where you work as a senior researcher?

Even before joining Avast, I have been interested in applications of game theory in network security. Network administrators and hackers are “playing games” against each other in computer networks from their inception. Furthermore, with the increasing importance of computer networks in all aspects of our lives, the games became very high stakes. In Avast, I am trying to apply the knowledge of game theory to create AI tools that would be robust against manipulation by our adversaries and, at the same time, understandable for our analysts. 

Our primary goal is to improve malware detection, but we expect our results to be applicable to many other tasks in Avast products and internal systems.

As you know, this year’s CyberSec&AI Connected will examine critical issues around AI for privacy and security. What aspects of this theme are you looking forward to discussing with other speakers and panelists at the conference?

One of the aspects of applying AI in security domains that I am interested in is the explainability of AI systems. I am looking forward to discussing what sort of explanations of decisions behind AI systems are acceptable for domain experts, customers, or regulators. 

Your talk is entitled ‘Playing poker with cybercriminals’. Can you give us an overview on what you’ll be discussing? 

In the talk I plan to briefly summarize the recent progress in creating AI systems that outperformed top players in various games. Afterwards, I will talk about the similarities and differences between these recreational games and computer security problems. Finally, I would like to share the vision and the initial results of our team at Avast.

Leaving aside your presentation, what are some of the more recent trends and developments around privacy that have caught your eye?

Regulators in developed countries understand the importance of privacy and come up with complex rules to try and ensure it is maintained. However, users very often do not understand the value and consequences of sharing their private information. It seems that recently this disconnect is growing even further and I am curious about how it will develop. 

What do you think can be done to increase awareness and knowledge among the general public and/or businesses in relation to privacy?

This is a very difficult problem. I would expect people to learn a lot from stories of their friends and relatives, who may have gotten into huge problems, because of some personal information leaks. It is difficult for people to imagine all these abstract risks, unless there are personal stories and examples attached to them. I believe that the mass media could help in this, but it is very difficult to do it without harming the privacy of the victims even further. I am afraid that the situation will just have to get much worse and, unfortunately, means more people will likely have to get first-hand experience before things can get better.

Due to current world events, this year’s conference will be done a bit differently. CyberSec&AI will be going virtual, connecting attendees wherever they are in the world. What excites you about this format and the opportunities it brings?

I like how accessible online conferences are for people from all over the world, regardless of family or budget restrictions limiting their opportunities to travel. I hope that many more people will be able to attend the conference and it will inspire people as it did last year, and create many new collaborations.

To hear Viliam present at CyberSec&AI Connected, book your place here. Bookings made before June 30 will qualify for our early bird discount and save you up to 50% off the full ticket price.

This article features

Viliam Lisy

Senior Research Scientist at Avast and Assistant Professor at CTU

Avast & Czech Technical University in Prague

Latest news

The positive development of a privacy preserv...

Since the beginning of the year, we have watched COVID-19 cases fluctuate around the world, turning ...

How AI can learn to win the cybersecurity “game”

Cybersecurity is an incredibly serious business. So thinking of it as a game may seem a little odd. ...

Global audience makes first virtual CyberSec&...

The first virtual edition of CyberSec&AI Connected took place on October 8th. Building on 2019...

CyberSec&AI Connected is here!

After a year of planning and preparation, CyberSec&AI Connected has arrived. Today sees delegate...