Track: EDA, MACHINE LEARNING
Time: 8:45 a.m. to 5:00 p.m.
Location: Room 3024
Machine learning is going full steam across several industries. Due to massive processing power and architectures that are honed to train large datasets, applications once thought to be infeasible are targeted now by machine learning algorithms. EDA has started to look into this area and utilize some of these ideas, at least on paper. There is a nice contrast though. EDA is a domain that is shaped with heuristics and tens of other algorithms and mathematical methodologies. Can Machine Learning claim its own territory for certain EDA applications? Can existing algorithms be sped up by using machine learning specific architectures instead? We think that bringing in both proponents of Machine Learning and traditional incumbents together, we can discuss what could be some best research and development directions for EDA in this area. We also plan to bring in people from external machine learning communities to get their view on the applicability to problems in our domains. We plan to build on these dynamics and educate the attendees on machine learning in general as well. The workshop consists of invited talks, panels, tutorials, and posters.
Manish Pandey, Synopsys, Mountain View, CA
Valeria Bertacco, University of Michigan, Ann Arbor, MI
Duane Boning, Massachusetts Institute of Technology, Cambridge, MA
Norman Chang, ANSYS, San Jose, CA
Soha Hassoun, Tufts University, Medford, MA
Andrew Kahng, University of California, San Diego, La Jolla, CA
David Pan, University of Texas at Austin, Austin, TX
Jinjun Xiong, IBM Corp., Yorktown Heights, NY