Neural Computation & Signal Processing Lab (NCSP)

Advanced Research Seminar

סמינר מתקדם במדעי המחשב

0368-5100-01  Semester I/II

Wednesday 14-16  Schreiber 309

Prof. Nathan Intrator

  Next Meeting: Wednesday, Oct 27, 2010 - Schreiber 309, 2:00pm-3:30pm

Using Iterative Ridge Regression to explore associations between Conditioned Variables

Nimrod Bar-Yaakov, Zehava Grossman, Nathan Intrator

Abstract

Bayesian and Dependency networks graphically map the joint probabilities of random variables. This paper addresses a specific case of joint probability mapping, where the information presented is the probabilistic associations of the random variables under a certain condition variable. The information is presented using a Directed Graph, where its nodes represent the random variables and its arcs the associations that arise under the condition.

Providing such mapping, may help uncovering associations that occur due to specific events, such as correlations between genetic mutations following medical treatment.

We introduce a method that efficiently handles the correlation mapping. This method is based on ridge-regression, where we can utilize a numerical robust and computational efficient algorithm implementation.

 

Seminar Overview

The seminar time will be used for a meeting of my lab students or prospective students who are interested in the field.

We shall have formal presentations as well as informal discussions.

Current applications include:

            Brain Imaging: EEG, fMRI – Epilepsy detection, network discovery, brain state interpretation

            Biomedical Signal Processing: Cardiac functionality interpretation

            Earthquake Detection

Financial Markets Anomaly Detection

Presentations should emphasize the machine learning, robust statistics and robust modeling in their work.

 

Nathan’s office hours: Schreiber 221, x7598,  Wednesday 4-5 (please set up by email)

 

 

NCSP Past seminars   2004-5 Sem I   2004-5 Sem II    Brain Imaging  2007-8   2008-9

Other relevant seminars:   PBC Seminar  Eshel Ben Yaakov  Eytan Ruppin  Ron Shamir  Hezy Yeshurun

 

   General instructions for seminar presenters

The presentation should have an introductory component that can enable all students understand the background of the seminar. They should then have a methodological component which explains at least a single method that can be used in a variety of applications. Finally, there should be some application results which demonstrate the usefulness of the proposed methods.

In contrast, a review presentation should describe several computational methods which are aimed at addressing a specific problem, together with a clear background of that problem. Preferably some comparison between the methods should be provided.

Abstract of the presentation should be sent to me up to three days before the presentation and a slides up to a day before the presentation.

 

           

Relevant Reading Material

 

Sound analysis

Auditory display of hyperspectral colon tissue images

Singing the Mind Listening

Sound features

Chris Raphael Rhythm changes

 

Biomedical signals and sensors

Robust measurement of Carotid Heart sound delay

Heart Mechanical and Electrical System

Segmentation of EKG signals

EKG Overview

Heart info and abnormalities (video)

 

Sensors

Cheap off-the shelf TinyOs operated robots

PicoRadio: Low power wireless node with sensors

Sensors Magazine

Xbow sensors

Machine learning and Statistics

Information theory T. Cover

Max Entropy Methods  R. Skiling

Pattern recognition and neural networks B. Ripley

Neural networks for pattern recognition Bishop

Digital Signal Analysis: A Computer Science

Perspective J. Stein.

Biomedical Signal Analysis R. M. Rangayyan 

Breath Sounds Methodology N. Gavriely

Introduction to Bayesian Networks K. Murphy

 

Software

Max/MSP Multimedia creation

TinyOS  operating sys for wireless applications

 

 

Next Seminars

            Alexandra Dana

            Sadna presentations

Past Presentations

 Date

Title

Speaker

Oct 28

Antagonistic relationship between gamma power and visual evoked potential revealed in human visual cortex

Eran Privman

Nov 11

Measurement of Cardiac Acoustic Delay

Alexandra Dana

Nov 25

Accurate Measurement of Echo Returns Under the Ground

Sasha Apartsin

Dec 2

Discovery of Multiple-Level Heart Sound Connected with Physiological Variability

Sveta Kofman

Dec 9

Multi-view point correspondence using spatio-temporal similarities in video sequences

Rotem Litman

Feb 24

Organizational Meeting

Nathan/Hezy

Mar 3

MSc Thesis Presentation

Ilana Podlipski

Mar 25

Inference of Brain Mental States from EEG Single Trials

Yehudit Hasson

Apr 18

Canceled

May 5

HMM fMRI

Elad Tsur

May 19

Predicting Free Choices in Humans: an Electrophysiological Study

Omri Perez

May 26

Granger/Simulations

Shahar Jamshi

Jun 9

SP EEG Epilepsy

Yaron Ziner