Transformative Computer Systems and Architecture Research Lab (TeCSAR)

Transformative Computer Systems and Architecture Research Lab (TeCSAR)

With the continuous growth in the complexity of deep learning and machine learning algorithms, the demand for novel computer architecture and system-level design paradigms is a most. At TeCSAR, we focus on transformative design methodologies, system-level modeling and computer hardware design for emerging machine learning and deep learning algorithms. The general roadmap of TeCSAR is algorithm/architecture codesign by offering novel scalable design tools and methodologies to bring algorithm-awareness in hardware design and enable rapid architecture design space exploration and prototyping. At the same time, we develop novel domain-specific computer architecture to reconcile the hardware programming flexibility and execution efficiency. At TeCSAR lab, Faculty, researchers, and students enable high-performance energy-efficient execution of a broad range of data-intensive numerical problems including scientific computing, graph processing, computer vision, and deep learning.

Primary Faculty: Hamed Tabkhi

Collaborators: Arun Ravindran (ECE), Andrew Willis (ECE), Babak Parkhideh (ECE), Omid Shoghli (ETCM), James Conrad (ECE), Amirhossein Ghassemi (MEES) and Gunar Schirner (ECE, Northeastern University)

Location: EPIC 2143