Date/Time
Date(s) - 26/09/2023
3:30 pm - 5:00 pm

Location: University Hall (UH) 112

Speaker: Steven Xiaogang Wang (York University)

Title: Machine Learning and Statistical Method for Large and Non-stationary Biological Signals

Abstract: Large and Non-stationary and highly noisy biological signals present a significant challenge for machine learning and statistical methods. First, we present a novel transformation called Lehmer transform and establish a theoretical framework used to compress and characterize large volumes of highly volatile time series data. The proposed method is a powerful data-driven approach for analyzing extreme events in non-stationary and highly oscillatory stochastic processes like biological signals. We then present a general framework for analyzing such signals with focus on EEG measurements. Building on pseudo-differential operators, we apply time-variant transform which allows neural network to learn a convolution kernel that changes over time and location. We also present a hybrid convolutional network that integrates both complex and real-valued components with Fourier Transformation for EEG classification.

Refreshments with speaker at 3pm in HH 216