This article is from the archives of the UB Reporter.
News

Srihari wins major award

  • “Buffalo has played an interesting role in the history of machine learning.”

    Sargur Srihari
    SUNY Distinguished Professor, Department of Computer Science and Engineering
By BRIAN PETERS
Published: October 24, 2011

Sargur N. Srihari, SUNY Distinguished Professor in the Department of Computer Science and Engineering and director of the Center of Excellence in Document Analysis and Recognition (CEDAR), has won the 2011 International Conference on Document Analysis and Recognition (ICDAR) Outstanding Achievements award.

Srihari is being honored with the award for his outstanding and continued contributions to research and education in handwriting recognition and document analysis, and for his service to the community.

He recently traveled to Beijing to accept the award and serve as a keynote speaker at the conference, held bi-annually by the International Association for Pattern Recognition.

His talk, entitled “Probabilistic Graphical Models in Machine Learning,” focused on the design of computer programs that learn and are able to modify their behavior in an environment of constantly changing information. Without machine learning, many computers that deal with rapidly changing data would require constant reprogramming.

Machine learning is crucial in fields such as document analysis and recognition due to the difficulty of expressing perceptual images, such as handwriting, in algorithms that computers can understand. Many of the advancements in machine learning were developed locally.

“Buffalo has played an interesting role in the history of machine learning,” Srihari says. “The first generation of machine learning programs, known as perceptrons, was developed at Calspan Corporation, here in Buffalo, in the 1960s. Many second-generation machine-learning programs were enabled by postal data collected at the Buffalo post office by UB CEDAR students.”

CEDAR is one of the largest research centers in the world devoted to developing new technologies that can recognize and read handwriting. Research by Srihari, his colleagues and students at CEDAR that enabled machines to recognize and understand handwriting was at the core of the first handwritten address-interpretation system used by the U.S. Postal Service. The research later was applied to “reading” tax forms and to forensic handwriting analysis.

“The third generation of machine-learning programs, which are based on probabilistic graphical models, are what enable many of today's natural language applications,” Srihari says.

An example of these third-generation programs is text analytics engines, which are capable of extracting such information as names and organizations from text in a variety of languages, including English, Chinese, Russian and Urdu.

The ICDAR conference also presents an award to promising researchers under the age of 40. Three of Srihari’s former students have won the award, including Venugopal Govindaraju, SUNY Distinguished Professor of Computer Science and Engineering and director of the Center for Unified Biometrics and Sensors at UB, who received the award in 2001; Jonathan Hull, director of Ricoh Research in Palo Alto, Calif., who won in 1997; and Tin Kam Ho, head of statistics research at Lucent Bell Labs in Murray Hill, NJ, who won in 1999.

In addition to founding and directing CEDAR, Srihari is a fellow of the Institute for Electrical and Electronics Engineers (IEEE) and the International Association of Pattern Recognition. He organized the first-ever international workshop on computational forensics in the U.S., which was held at the National Academy of Sciences. He also has been a member of federal committees charged with developing and applying the best scientific standards to important social issues, serving on the Board of Scientific Counselors of the National Library of Medicine and the National Academy of Sciences Committee on Identifying the Needs of the Forensic Science Community. In addition, he is the author of more than 300 peer-reviewed publications and the subject of major coverage in the news media.