Machine learning has long been linked to Physics as both fields deal with exponentially large spaces and complex cost-function landscapes. Physics intuition allows to deeper understand underlying processes in Neural Networks and mechanics of various empirically invented tricks. In my talk I will outline how physics-inspired approaches bring us closer to less redundant and more universal machine learning models.
Mykola Maksymenko is a Research Lead at SoftServe where he works on applied ML and complexity-intense problems. These can range from biosignal recognition applications to assessing self-organisation in financial ecosystems. Mykola holds a PhD in theoretical physics and worked as a researcher at Max-Planck Institute for the Physics of Complex Systems and Weizmann Institute of Science studying dynamics of quantum states, complex networks and exotic magnetic materials. He is currently interested in overlaps between physics and machine learning which could lead to deeper insights and useful practical applications on both sides.
Event Timeslots (1)
Track B (Lower Floor)