Andrei PĂUN

• Andrei received his BSc in Mathematics with a major in Computer Science from University of Bucharest (Romania) and later he received a Master’s followed by a PhD in Computer Science from University of Western Ontario (Canada) in 2003. His PhD supervisor was Dr. Sheng Yu, from whom he “inherited” the interest in Finite Automata and OOP modeling as well as the Biocomputing research field that was the main thrust of his PhD dissertation. Following the PhD Andrei was a postdoc at Technical University of Wien (Austria) and later at University of Rovira i Virgili (Spain). He then moved to the US as an assistant and later associate professor of Computer Science (tenured) at Louisiana Tech University (USA). He later returned to Europe having positions at University of Bucharest (Romania), National Institute of R&D for Biological Sciences (Romania) and recently he accepted a position of director of the Research Institute for Artificial Intelligence "Mihai Draganescu" of the Romanian Academy. His main interests are Biocomputing, Bioinformatics, AI as well as finite automata and other related areas.

Computing with Artificial Neural Networks: SNP systems

• Following the recent Nobel prize awards in Physics and Chemistry we will give a novel view of the AI field as well as other models of computing inspired from the working of neurons: Spiking Neural Systems (SNP). The SNP systems are a type of distributed, parallel, neural-like computation model inspired by how biological neurons process information and communicate through spikes. In the past decade, our understanding of neurons and the brain has advanced significantly, largely due to the Human Brain Project in Europe and the BRAIN Initiative in the US. This discussion will highlight findings related to SNP systems and their variants as computational models, many demonstrating Turing Universality under certain constraints. We will also examine recent research and ongoing questions, particularly concerning Spiking Neural P systems with communication on request (SNQ P systems). Notably, we have designed compact, Turing universal SNQ P systems with a limited number of neurons, including a generative SNQ P system that performs acceptance and function computation using just 12 neurons (11 for acceptance and 21 for function computing).