Rapid advancement of machine learning make it possible to consider large amounts of data to learn from. In most of the implementations of reinforcement learning facing this type of data, approximation is obtained by neural networks and the process of drawing information from data is mediated by a short-term memory that stores the previous experiences for additional re-learning, to speed-up the learning process, mimicking what is done by people. In this work, we are proposing a range of novel computational approaches able to selectively filter the informational or cognitive load for the agent’s short-term memory, thus simulating the attention mechanism characteristic also of human perception. Using the proposed attention ﬁlter block architecture, we were able to devise a variety of frameworks of agent’s perception that are able to adapt to its environment by selecting the most suitable experiences. The adaptation also resulted in an emergence of different behavioral characteristics or traits among artificial learning agents.
June 10, 2019 (Monday)
Room 205, Building E, Karlovo nám. 13
After his studies in Bosnia and Herzegovina Mirza got his PhD at the Politecnico di Milano in Italy. His thesis was about perception as a behavior inducing mechanism - the topic which he will introduce in his talk. Apart from programming and web development he is interested in research in the field of AI, machine learning, multi-agent systems and affective computing. Currently, Mirza is applying to become a post doctoral student at our department.