EE Seminar: Learning From Multi-View High Dimensional Data
Speaker: Ofir Lindenbaum
Ph.D. student under the supervision of Prof. Arie Yeredor and Prof. Amir Averbuch
Wednesday, June 28h, 2017 at 15:00
Room 011, Kitot Bldg., Faculty of Engineering
Learning From Multi-View High Dimensional Data
Abstract
High dimensional “big data” is often encountered in various fields, such as audio analysis, video analytics and data mining. One of the associated challenges is how to extract meaningful information from such data, which is generally difficult to analyze as is. Dimensionality reduction methodologies reduce the size (dimension) of objects in the dataset while preserving the coherence of the original data, such that clustering, classification, manifold learning and many other data analysis tasks can be applied in the reduced space.
We consider learning a reduced dimensionality representation from datasets obtained under multiple views. Such multiple views of datasets can be obtained, for example, when the same underlying process is observed using several different modalities, or measured with different instrumentation. Our goal is to effectively exploit the availability of such multiple views for various purposes, such as non-linear embedding, manifold learning, spectral clustering, anomaly detection and non-linear system identification.
Our proposed method exploits the intrinsic relation within each view, as well as the mutual relations between views. We do this by defining a cross-view model, in which an implied Random Walk process between objects is restrained to hop between the various views. In this talk I will describe two frameworks for multi-view dimensionality reduction. Applications for manifold learning, clustering, classification and detection of seismic events will be presented.
