CCSP Seminar: Sina Miran, Dec. 7 (Thursday, 5 PM)

Thursday, December 7, 2017
5:00 p.m.-6:30 p.m.
AVW 2168 (ISR Conference Room)
Ajaykrishnan Nageswaran
301 405 3661

Title:  Real-Time Tracking of Selective Auditory Attention from M/EEG: A Bayesian Filtering Approach
Abstract: Humans are able to identify and track a target speaker amid a cacophony of acoustic interference, which is often referred to as the cocktail party phenomenon. Results from several decades of studying this phenomenon have culminated in recent years in various promising attempts to decode the attentional state of a listener in a competing-speaker environment from non-invasive neuroimaging recordings such as magnetoencephalography (MEG) and electroencephalography (EEG). To this end, most existing approaches compute correlation-based measures by either regressing the features of each speech stream to the M/EEG channels (the decoding approach) or vice versa (the encoding approach). These procedures operate in an offline fashion, i.e., require the entire duration of the experiment and multiple trials to provide robust results. Therefore, they cannot be used in emerging applications such as smart hearing aid devices, where a single trial must be used in real-time to decode the attentional state. 

In this talk, we will develop an algorithmic pipeline for real-time decoding of the attentional state. Our proposed framework consists of three main modules: 1) Real-time and robust estimation of encoding or decoding coefficients, achieved by sparse adaptive filtering, 2) Extracting reliable markers of the attentional state, and thereby generalizing the widely-used correlation-based measures thereof, and 3) Devising a near real-time state-space estimator that translates the noisy and variable attention markers to robust and reliable estimates of the attentional state with minimal delay. Our proposed algorithms integrate various techniques including forgetting factor-based adaptive filtering, L1-regularization, forward-backward splitting algorithms, fixed-lag smoothing, and expectation maximization. We validate the performance of our proposed framework using comprehensive simulations as well as application to experimentally acquired M/EEG data. Our results reveal that the proposed real-time algorithms perform nearly as accurate as the existing state-of-the-art offline techniques, while providing a significant degree of adaptivity, statistical robustness, and computational savings.

Audience: Graduate  Faculty  Post-Docs 


February 2020

26 27 28 29 30 31 1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
1 2 3 4 5 6 7
Submit an Event