Leonard E. Baum

from Wikipedia, the free encyclopedia

Leonard "Lenny" E. Baum (born August 23, 1931 in Brooklyn , New York , † August 14, 2017 in Princeton , New Jersey ) was an American mathematician . He was a pioneer in Hidden Markov Models (HMM).

Baum studied mathematics at Harvard University with a bachelor's degree in 1953 ( summa cum laude ) and a doctorate in 1958 ( Derivations in commutative semi-simple Banach algebras ). He spent several years as a post-doctoral student at the University of Chicago . His pioneering work on HMM came in the late 1960s when he was at the Institute for Defense Analyzes (IDA), a Princeton think tank where he was from 1959 to 1978. The work was related to research on automatic speech recognition. The Baum-Welch algorithm for HMM, which made the first effective speech recognition programs possible, is named after him and Lloyd R. Welch . Most of his work at IDA was secret. According to Baum's obituary in the New York Times, he was seen there as a persistent and outstanding problem solver with diverse interests (as a "Renaissance man" ) and as a patient mentor and teacher.

He later worked for hedge funds on Wall Street as a financial mathematician, including Monemetrics (a forerunner of Renaissance Technologies), where he was brought in 1977 by founder James Simons, who had also worked at IDA (as a code breaker during the Vietnam War ) and therefore knew him. He applied there, among other things, HMM. However, he withdrew early after his eyesight deteriorated significantly. But that didn't stop him from traveling around the world and until shortly before his death he hiked around four miles a day. He pursued mathematical research until shortly before his death.

His parents were Sophia Fuderman and Morris Baum. In 1953 he married Julia Lieberman and had two children.

Individual evidence

  1. ^ Obituary in the New York Times , accessed January 14, 2018
  2. ^ IEEE Global History, Hidden Markov Model
  3. ^ PhDs in Mathematics from Harvard
  4. At IDA, after Baum's obituary in the New York Times, he is assigned the institute's unofficial motto: No idea is bad. A bad idea is good. A good idea is terrific .
  5. LE Baum, T. Petrie, G. Soules, N. Weiss A maximization technique occurring in the statistical analysis of probabilistic functions of Markov chains , Ann. Math. Statist., Vol. 41, 1970, pp. 164-171. Baum published about it as early as 1967.
  6. Article on Simons at Bloomberg, 2007