У оквиру традиционог колоквијума Института за физику, у среду, 17. априла 2019. у 12 часова у сали „Звонко Марић“ предавање под насловом
Machine Learning and Condensed Matter: What can we learn?
ће одржати проф. др Никола Рењо (Laboratoire de Physique, Ecole Normale Superieure – CNRS, Paris, France).
Relying on the advance of computing power, algorithms and large databases, machine learning (ML) has become a paradigmatic change in many areas. The versatility of the techniques and the success of ML have triggered a large interest in this approach applied to condensed matter. Obviously, it can be used as an advanced data analysis method for experimental results. In this talk, we will focus on ML in more theoretical contexts. Through a few examples, we will unveil its potential applications. We will discuss how it can discriminate between different phases in numerical simulations and how it can be used to encode many-body quantum states in an efficient manner.