Campus news

Computer scientists to cheer for machine at "Jeopardy!" viewing party

By Sarah Yang

WHAT

Viewing party at the University of California, Berkeley, for the long-anticipated showdown pitting “Jeopardy!” game show champions against Watson, an IBM supercomputer. The three-day man vs. machine matchup, some seven years in the making, began airing yesterday (Monday, Feb. 14).

The public viewing of Wednesday’s finale, hosted by the Center for Information Technology Research in the Interest of Society (CITRIS), will lead off with a discussion by leading computer scientists about how “Jeopardy!” showcases the capabilities of artificial intelligence and machine learning.

WHEN

6:15-7:30 p.m., Wednesday, Feb. 16. Sastry and Simon will kick off the event at 6:15 p.m., followed by the airing of the final episode from 7-7:30 p.m.

WHO

The pre-game festivities will be led by:

Shankar Sastry, dean of UC Berkeley’s College of Engineering. Sastry has made his mark in a number of computing research fields, including artificial intelligence, autonomous robotics, sensor networks, biological motor control and cybersecurity. He has held many leadership positions at UC Berkeley, including as director of CITRIS and the Banatao Institute@CITRIS Berkeley.

Horst Simon, deputy director of Lawrence Berkeley National Laboratory. Simon is an internationally recognized expert in computer science and applied mathematics. He is founding director of Berkeley Lab’s Computational Research Division, and co-editor of the biannual TOP500 list that tracks the most powerful supercomputers worldwide, as well as related architecture and technology trends.

WHERE

Banatao Auditorium, 3rd floor of Sutardja Dai Hall. Seating is first come, first served. Public inquiries can be sent to citriseditor@citris-uc.org.

DETAILS

Representing the human brain are Jeopardy’s two best champs: Ken Jennings, who had the game show’s longest winning streak, and Brad Rutter, the show’s top money winner. In the machine corner stands Watson, a supercomputer equipped with custom artificial intelligence algorithms designed to answer natural language questions with a single, precise answer.

According to IBM, a single-processor computer could take up to two hours to answer a single question that a typical “Jeopardy!” contestant can answer in less than three seconds. To rival its human competitors, Watson employs 90 servers and nearly 3,000 POWER7 computer cores working in a massively parallel system.

As a possible precursor of what’s to come, Watson emerged victorious in a three-category “Jeopardy!” practice round last month.

IBM’s website has more background on Watson’s journey to become a “Jeopardy!” contestant.