My (here’s a short, glitzy version of my CV) research studies (see list of my publications) different incarnations of Machine Learning (ML), and various applications of those. My interests are found in theoretical, algorithmic and application-oriented aspects of the question: ‘What makes an ML algorithm work for a given case?’ This is a non-applied question really, but applications play a fundamental role here.
It is fair to say that ML (or data-based techniques) are evolving partly due to the ever-increasing availability of computational power. But while the availability of fast hardware is mostly regulating the latter, the limits of the former are often dictated by the availability of proper algorithms. A large class of problems do not require massive hardware resources. Properly designed algorithms are (1) theoretically sound, (2) in-tune with its intended application aim, and (3) computationally attractive. These topics are invariantly present underlying trending topics as neural networks, kernel machines, compressed sensing, deep learning, business analytics and Big Data.
A particular focus of my recent research is to address the interplay between ML and cybersecurity via realtime ML and worst-case analysis. Many of the industrial/academic challenges of ML are rooted in the lack of reliability of many of the techniques. This is a fertile and open-ended research track that finds more and more resonance in industrial reality as well.