What Do Sea Slugs, People and Computer Circuits Have in Common?

REDMOND and SEATTLE, Wash., Aug. 17, 1998 — The secret to intelligent computing as well as clues to understanding how animals and humans learn may lie in joint studies of biological organisms, such as sea slugs, and computer programs. The 1998 University of Washington (UW) and Microsoft Research Summer Institute on Intelligent Systems, Aug. 18-23, will bring together some of the finest researchers in these fields to explore principles of intelligent behavior in humans and animals and how they might be applied in computer science.

This meeting is an assembly of a diverse group of computer scientists and biologists to discuss how findings about nervous systems can help computer scientists build smarter systems and how insights from computer science can assist with research in neurobiology. The Summer Institute will bring together 45 of the top researchers in biology and computer science to study the fundamental questions about intelligent systems from their diverse perspectives. Marvin Minsky, co-founder of the MIT Artificial Intelligence (AI) Lab and pioneer in AI research, will present the meeting charge. Other participants include UW MacArthur Award winner Tom Daniel, who will present research on intelligent sensing and control of complex motor systems in animals.

“The neurons in our brains don’t work like computer chips,”
said Chris Diorio, UW assistant professor of computer science and engineering and co-organizer of the Summer Institute.
“And our brains don’t work like a digital computer. They aren’t very good at math, but they do things that are very hard for a computer to do, like understanding speech. If we can figure out how the neurons in our brains represent information and learn, then perhaps we can copy biology and build new types of computer chips with broader capabilities.”

Now in its second year, the Summer Institute was jointly created by the UW computer science and engineering department and Microsoft Research. Its goal is to identify an emerging area of computer science and bring together top researchers from diverse fields to chart new territory. UW and Microsoft partners aim to spur dialogue and interdisciplinary research to accelerate progress in developing promising new technologies.

“Nervous systems may hold insights for computer scientists on principles for building robust, intelligent systems,”
said Eric Horvitz, senior researcher at Microsoft Research and co-organizer of the Summer Institute.
“Exploring biological solutions to intelligence is akin to the early studies of birds in flight, which provided the initial glimmers of understanding that eventually blossomed into principles of aerodynamics. On the flip side, computer scientists have developed theoretical models of learning and decision-making that may be useful for guiding neurobiology research.”

Computer scientists are eager to understand how nervous systems might offer useful clues for new ways of computing. The decision theory and adaptive systems group at Microsoft Research has been pursuing principles of intelligence – work that has already led to practical applications in consumer software. The group’s interests include understanding how computers can integrate information from multiple sources, allocate limited computational resources to different tasks, and take intelligent actions amid the uncertainties of real-world environments.

Biologists are indeed looking for clues from computer scientists to better understand how nervous systems function to produce intelligent behavior in animals and humans. Researchers at Friday Harbor Laboratories can map chemical and electrical signals in the nervous systems of sea slugs and observe resulting behaviors, but are often stymied on how the signaling and behaviors are linked.

“We can study the circuitry and the behavior, but it isn’t always clear what the details of the connections are,”
said Dennis Willows, director of Friday Harbor Labs and the third organizer of the Summer Institute.
“If biologists can talk to engineers and scientists who have developed models or circuits that function like neural systems in living organisms, I suspect we will learn of several possible connections as well as ways to test those connections.”

The University of Washington, with 34,000 students and 3,500 faculty members, is one of the nation’s premier research universities. UW is consistently among the top five institutions in annual federal research grants and can claim over 100 members of national academies, eight MacArthur Foundation Award winners and four Nobel Prize winners in the past decade. The UW department of computer science and engineering, established in 1967, is ranked among the top 10 in the nation for its caliber of research and teaching.

Founded in 1991, Microsoft Research is dedicated to conducting both basic and applied research in computer science and software engineering. Microsoft Research started with a handful of researchers and has grown steadily to include more than 300 computer scientists and engineers in a wide variety of areas including decision theory, speech technology, databases, user interface and 3-D graphics. Although less than a decade old, Microsoft Research is already known widely as one of the best computer science research laboratories in the world.

Founded in 1975, Microsoft (Nasdaq
) is the worldwide leader in software for personal computers. The company offers a wide range of products and services for business and personal use, each designed with the mission of making it easier and more enjoyable for people to take advantage of the full power of personal computing every day.

Microsoft is either a registered trademark or trademark of Microsoft Corp. in the United States and/or other countries.

Other product and company names herein may be trademarks of their respective owners.

More information about Summer Institute can be found on the Web at http://isys.microsoft.com/ and at http://www.microsoft.com/presspass/ . For Microsoft Research information, visit http://research.microsoft.com/ .

Related Posts