Madhu Sudan Receives Nevanlinna Prize
On August 20, 2002, the Sudan has been a main contributor to the devel- Rolf Nevanlinna Prize was opment of the theory of probabilistically checkable awarded at the opening cer- proofs. Given a proof of a mathematical statement, emonies of the International the theory provides a way to recast the proof in a Congress of Mathematicians form where its fundamental logic is encoded as a (ICM) in Beijing, China. The sequence of computer bits. A “verifier” can, by prizewinner is MADHU SUDAN. checking only some of the bits, determine with In 1982 the University high probability whether the proof is correct. What of Helsinki granted funds is extremely surprising, and quite counterintuitive, to award the Nevanlinna is that the number of bits the verifier needs to Prize, which honors the examine can be made extremely small. The theory work of a young mathe- was developed in papers by Sudan, S. Arora, U. Feige, matician (less than 40 years S. Goldwasser, C. Lund, L. Lovász, R. Motwani, of age) in the mathematical S. Safra, and M. Szegedy. For this work, these authors aspects of information sci- jointly received the 2001 Gödel Prize of the Associ- Madhu Sudan ence. The prize is pre- ation for Computing Machinery. sented every four years in Also together with other researchers, Sudan has conjunction with the ICM. Previous recipients of the made fundamental contributions to understanding Nevanlinna Prize are: Robert Tarjan (1982), Leslie the nonapproximability of solutions to combina- Valiant (1986), Alexander Razborov (1990), Avi torial optimization problems. This work connects Wigderson (1994), and Peter Shor (1998). to the fundamental outstanding question in The Nevanlinna Prize is awarded by the theoretical computer science: Does P equal NP? A International Mathematical Union on the advice of problem is in P if there is a polynomial-time algo- a selection committee. The selection committee rithm that solves it. A problem is in NP if it has the for the 2002 prize consisted of: Andrei Agrachev, property that a proposed solution can be checked Ingrid Daubechies, Wolfgang Hackbusch, Michael O. in polynomial-time but that no polynomial-time Rabin (chair), and Alexander Schrijver. algorithm is known to solve it. What Sudan and Madhu Sudan was born on September 12, 1966, others showed is that, for certain combinatorial in Madras (now Chennai), India. He received his optimization problems, approximating an optimal B. Tech. degree in computer science from the In- solution is just as hard as finding an optimal dian Institute of Technology in New Delhi (1987) solution. This result is closely related to the work and his Ph.D. in computer science at the University on probabilistically checkable proofs. of California at Berkeley (1992). He was a research The third area in which Sudan made important staff member at the IBM Thomas J. Watson Re- contributions is error-correcting codes. A class of search Center in Yorktown Heights, New York widely used codes is the Reed-Solomon codes (and (1992–7). He is currently an associate professor in their variants), which were invented in the 1960s. the Department of Electrical Engineering and Com- For forty years it was assumed that the codes could puter Science at the Massachusetts Institute of correct only a certain number of errors. By creat- Technology. ing a new decoding algorithm, Sudan demonstrated Madhu Sudan has made important contribu- that the Reed-Solomon codes could correct many tions to several areas of theoretical computer sci- more errors than previously thought possible. ence, including probabilistically checkable proofs, —Allyn Jackson nonapproximability of optimization problems, and error-correcting codes.
1266 NOTICES OF THE AMS VOLUME 49, NUMBER 10