Mari Ostendorf named RSE Fellow
System Design Methodologies Professor Mari Ostendorf joins 63 new RSE fellows, announced from across sciences, arts, education, business and public life.
Overview
Data sciences are fundamentally transforming nearly every area of engineering, science, and society. The University of Washington’s Electrical & Computer Engineering faculty are making fundamental contributions to many different areas of data sciences, including machine learning, AI, optimization, information theory, computer vision, and speech and natural language processing. Many of our data sciences faculty hold secondary appointments in applied mathematics, computer science and engineering, bioengineering, and other departments, and are active participants in cross-disciplinary institutes such as UW’s eScience Institute, the Allen Institute of Artificial Intelligence and the Bloedel Hearing Research Center.
Topics
Machine Learning
Artificial intelligence (AI), mathematical optimization and information theory.
Faculty: Katrin Kirchhoff, Jeffrey A. Bilmes, Les Atlas, Maryam Fazel, Sreeram Kannan, Mari Ostendorf, Ming-Ting Sun, Eli Shlizerman, Jenq-Neng Hwang, Linda Shapiro, Hannaneh Hajishirzi, Shwetak Patel, Radha Poovendran
Statistical Signal Processing
Theory, algorithms, signal processing systems and signal processing applications (i.e. biomedical, geophysical signals and synthetic signals).
Faculty: Ming-Ting Sun, Les Atlas, Mari Ostendorf, Brian A. Nelson, Jenq-Neng Hwang
Speech and Natural Language Processing
Speech recognition, natural language understanding, computational linguistics and web-based language techniques.
Faculty: Mari Ostendorf, Katrin Kirchhoff, Jeffrey A. Bilmes, Hannaneh Hajishirzi, Les Atlas
Computer Vision and Image Processing
Video analysis, surveillance, object recognition, activity recognition, medical image analysis and video compression
Faculty: Ming-Ting Sun, Jenq-Neng Hwang, Linda Shapiro, Jeffrey A. Bilmes, Eve Riskin, Radha Poovendran
Latest News
GIX student team "SnapSort!" advances to finals of Microsoft Imagine Cup
ECE lecturer John Raiti's MSTI students created SnapSort!, an intelligent, real-time trash sorting assistant that uses computer vision and machine learning.
The one ring — to track your finger’s location
Could this be the beginning of a new era of VR and AR tech? ECE doctoral student Farshid Salemi Parizi and Prof. Shwetak Patel have created AuraRing, a 5 DoF electromagnetic ring and wristband combination that can detect the precise location of someone’s finger and continuously track hand movements.
Jennifer Mankoff and Eve Riskin lead ECE's co-sponsored UW EXP study
The UW EXP study is a biometric approach to understanding and addressing student wellness.
Ostendorf wins 2018 Flanagan award
For innovative research
Maryam Fazel to lead NSF TRIPODS+X in data science
Project will enhance the successful "hack week" model
Researchers
- Sreeram Kannan
- Eli Shlizerman
- Shwetak N. Patel
- Radha Poovendran
- Ming-Ting Sun
- Linda G. Shapiro
- Eve A. Riskin
- Mari Ostendorf
- Brian A. Nelson
- Katrin Kirchhoff
- Jenq-Neng Hwang
- Hannaneh Hajishirzi
- Maryam Fazel
- Jeffrey A. Bilmes
- Les Atlas
Research Labs
- MachinE Learning, Optimization, and Data Interpretation (MELODI) Laboratory
- Transformation, Interpretation and Analysis of Language (TIAL)
- Intelligent Systems Lab
- Graphics and Imaging Lab
- Data-Driven Dynamical Systems
- Signal, Speech and Language Interpretation Lab
- Silicon System Research Lab
- Ubicomp (Ubiquitous Computing) Research Lab
- Data Compression Lab
- Design, Test and Reliability Research Laboratory
- Information Processing Lab
- Information Theory Lab
- Interactive System Design Laboratory
Latest Projects
- simsong.org
- Digital Pathology: Accuracy, Viewing Behavior and Image Characterization (with PI: Joann Elmore at Harborview and others)
- 3D Head Reconstruction from Images or Videos (with Ira Kemmelmacher-Shlizerman in CSE)
- Expression Recognition with Deep Neural Nets (with Barbara Mones in CSE)
- Automatic Recognition of Power Line in Millimeter Wave Radar Video
Representative Publications
- Surface Light Field Compression Using a Point Cloud Codec
- A Data-driven Point Cloud Simplification Framework for City-scale Image-based Localization
- Full-Capacity Unitary Recurrent Neural Networks
- Deep Submodular Functions: Definitions and Learning
- Exploiting Tradeoffs for Exact Recovery in Heterogeneous Stochastic Block Models
- Y-Net: Joint Segmentation and Classification Diagnosis of Breast Biopsy Images