DrS on Physics-Mathematics, Professor

Oleksandr Letychevskyi

Affiliation:

• Theory Digital Automata Department,

the Glushkov Institute of Cybernetics,

National Academy of Sciences of Ukraine

• Heriot-Watt University, Edinbourgh, United Kingdom

Title: NEURO-SYMBOL APPROACH IN CYBER SECURITY

Abstract:

The use of neural network technology in cyber security is complicated by many problems, such as the lack of relevant data sets, false positive detection of intrusions into the network environment, adversarial attacks, and others. To overcome these challenges, it is possible to use the combined methods of using neural networks and the algebraic approach, which is known in literature as the neuro-symbolic approach. The report considers the possibility of using so-called cognitive networks that use the strengths of both approaches. The capabilities of this technology in detecting and preventing intrusions into the computer and network environment will be presented.

Short-bio:

Oleksandr. Letychevskyi is a head of Theory Digital Automata Department at the Glushkov Institute of Cybernetics (Kyiv, Ukraine). He has worked as a contractor of Motorola and as a researcher at Missouri University (Rolla, USA). His research interests include algebraic approaches and neuron network technology for cyber security, verification, model-driven development, and testing of safety critical systems. It includes research in blockchain, biology, software and hardware development subject domains.