Dr. Alexandros Iosifidis

Department of Engineering, ECE, Aarhus University, Denmark  

 

Talk Title
Efficient Neural Network Models for Signal Processing Applications
Talk Abstract
 Recent advances in machine learning led to remarkable solutions in many research problems, notably in fields of computer vision, natural language processing and games. This is due to new parallel computing capabilities for scientific computing (Graphical Processing Units–GPUs and distributed computing), the availability of enormous sets of annotated data, and methodological contributions in a family of models called artificial neural networks, and commonly referred to as Deep Learning. Such state-of-the-art models are commonly formed by an enormous number (in the order of hundreds of thousand, or even millions) of parameters which are jointly tuned in an end-to-end optimization process to fit the training data. However, the use of these powerful models in several Signal Processing applications is still unrealistic due to their underlying memory and computation restrictions. In this talk, I will provide an overview of efficient neural network models proposed recently to highly reduce computations. Focus will be given to our bilinear neural networks, models exploiting multi-linear approximation of tensorial weights, and a recently proposed family of feedforward networks, the Generalized Operational Perceptrons, leading to compact network topologies by modeling various types of nonlinearities encoded in the data. Applications of these models in financial time-series and image analysis problems will be discussed. This talk will be relevant for both researchers and practitioners of neural networks, since it involves new state-of-the-art methodologies and ready-made solutions for a wide range of problems.
Short Biography
Alexandros Iosifidis is an Assistant Professor of Machine Learning & Computer Vision in Aarhus University, Denmark. His research interests lie in the areas of Machine Learning, Pattern Recognition, Computer Vision and Computational Finance. He has (co-)authored 50 articles in international journals, 71 conference papers, 4 book chapters, and one patent in topics of his expertise. His work has attracted 1300+ citations, with h-index of 18 (Google Scholar). Dr. Iosifidis is a Senior Member of IEEE. He served as an Officer of the Finnish IEEE Signal Processing/Circuits and Systems Chapter from 2016 to 2018. He is currently serving as an Associate Editor in Neurocomputing and IEEE Access journals, as an Area Editor in Signal Processing: Image Communication journal and he was an Area Chair in IEEE ICIP 2018. He was a TPC member/reviewer in more than 50 International Conferences focusing on Machine Learning methodologies and their applications in various topics. His work received several awards, including the H.C. Ørsted Forskerspirer 2018 prize for research excellence, the Academy of Finland Postdoc Fellowship, the best student paper awards in IPTA2016 and listed as best papers in VCIP2017, IPTA2016, IJCCI2014 and IEEE SSCI2013.
 
Talk Keywords
Neural Networks, Deep Learning, Multi-linear Neural Networks, Generalized Operational Perceptrons.
 
Target Audience
Students, Post doctoral, Industry, Doctors and professors
 
Speaker-intro video
TBA 
 

Sponsors & Partners