Dr. Sotirios Chatzis has recently published a paper in the IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI). IEEE TPAMI is the most prestigious journal in Machine Learning, and the most cited journal in Electrical and Computer Engineering. As a result, it is a highly selective journal only publishing groundbreaking papers in the field of machine learning.

In the published paper entitled "The Infinite-Order Conditional Random Field Model for Sequential Data Modeling," the author presents the first ever conditional random field (CRF) formulation capable of capturing infinitely-long time dependencies; that is, a CRF the transition potentials of which are non-Markovian, and depend on the whole history of transitions. The method outperformed all the state-of-the-art methods in several benchmarks, dealing with applications from as diverse fields as bioinformatics, speech processing, natural language processing, and handwriting recognition.