سمینار بینالمللی پژوهشکده مخابرات نظری و قطب علمی سیستمهای مخابرات
سخنران:
Dr. Hamed Nili
Postdoctoral Researcher
Oxford University
Department of Experimental Psychology
عنوان سمینار:
New insights about the nature of brain representations via pattern
similarity analysis
دوشنبه 3 مهرماه از ساعت 16:30 تا 18
دانشکده مهندسی برق، دانشگاه صنعتی شریف (برق 4)
Abstract:
Recent advances in techniques for analyzing neuroimaging data has shed light into the nature of brain representations. In this talk, I will review some recent work covering a range of topics from memory encoding and retrieval to concept formation. I will explain
how the paradigm shift from univariate analysis to multivariate pattern analysis of brain responses allows answering new questions
from brain measurements. Finally, I argue that many of the questions rely on the concept of pattern distances and that a particular approach to multivariate analysis is most appropriate for data analysis.
About Speaker:
Dr. Nili received his BSc degree in Electrical Engineering at Sharif University of Technology, specialized in Control engineering. His undergraduate project
was an attempt to design a hands-free wheelchair control system. There, he used features from surface EMG signals to classify four types of hand motions
based on Artificial neural networks. He then did an MSc on applied digital signal processing in Southampton. This was followed by two years of research
in Professor John Duncan's lab in the University of Cambridge MRC Cognition and Brain Sciences Unit. Being John's Research Associate, he worked on
single-cell data to study target detection in the prefrontal cortex. He then started his PhD in the same division at Cambridge under the supervision of Dr
Nikolaus Kriegeskorte. His PhD mainly consisted of developing methods for multivariate data analysis (e.g. the RSA toolbox) and also orientation
invariance in the human visual system (using fMRI). Currently, in his post-doc in the Summerfield lab at Oxford University, he is mostly using the methods
that he developed in his PhD as tools for studying human learning. His projects span a range from learning abstract categories to transfer learning (the
mechanisms by which learning a task would benefit our performance on similar tasks). He uses fMRI, EEG and behavioural analysis and modelling to study
the questions of interest.
