Richard E. Bellman

Richard Ernest Bellman was a highly influential American applied mathematician. He is most famous for introducing the concept of dynamic programming in 1953. This method provides a way to solve complex problems by breaking them into smaller, manageable parts. His work helped create new ways to understand decision-making and control systems. Bellman's ideas reached far beyond pure math, impacting fields like engineering, economics, and biology. He was a deeply productive scholar who changed how scientists approach optimization.

At the heart of his work is the Bellman equation. This is also known as the dynamic programming equation. It serves as a necessary condition for optimality in mathematical optimization. In simple terms, it helps find the best possible way to complete a task. If a problem can be solved using optimal control theory, it can often be solved using a Bellman equation. This equation was first used in engineering control theory. Later, it became a vital tool for economists trying to model complex systems.

There is also a more advanced version called the Hamilton–Jacobi–Bellman (HJB) equation. This is a partial differential equation used in the study of optimal control theory. The solution to this equation is called a "value function." This function tells you the optimal cost-to-go for a specific dynamical system. While the discrete-time version is called the Bellman equation, the HJB equation works in continuous time. This version can be seen as an extension of earlier work in physics by William Rowan Hamilton and Carl Gustav Jacobi. It can even solve classical problems like the brachistochrone problem.

Bellman also identified a major challenge in mathematics known as the curse of dimensionality. He coined this term to describe how problems grow harder as you add more dimensions. When you add extra dimensions to a mathematical space, the volume increases exponentially. This makes numerical solutions much more difficult for computers to handle. For example, sampling a 10-dimensional hypercube requires vastly more points than sampling a simple line. A 10-dimensional hypercube can be a factor of 10 to the 18th power larger than a unit interval. This massive increase in complexity requires significant computer time to solve.

Richard Bellman was born in New York City in 1920. His parents, Pearl and John James Bellman, operated a grocery store in Brooklyn. He attended Abraham Lincoln High School in 1937 and later graduated from Brooklyn College in 1941. He earned an MA from the University of Wisconsin and a Ph.D. from Princeton University in 1946. During World War II, he worked in the Theoretical Physics Division at Los Alamos. Starting in 1949, he spent many years working at the RAND corporation. It was during this time that he developed his groundbreaking theories.

Bellman was an incredibly prolific writer and researcher. Throughout his career, he published 39 books and 619 papers. Even after a difficult brain tumor diagnosis in 1973, his productivity remained high. Despite suffering from severe disabilities following his surgery, he published over 100 papers in his final 11 years. He was a professor at the University of Southern California and received many honors. These included being a Fellow of the American Academy of Arts and Sciences in 1975. He also joined the National Academy of Engineering in 1977 and the National Academy of Sciences in 1983. In 1979, he was awarded the IEEE Medal of Honor for his work on decision processes.

In his later years, Bellman shifted his focus toward biology and medicine. He called these fields the "frontiers of contemporary science." In 1967, he became the founding editor of the journal Mathematical Biosciences. This journal remains a leading publication in the field of mathematical biology today. To honor his legacy, the Bellman Prize in Mathematical Biosciences was established. This prize is awarded every two years to the best research paper in the journal. His work continues to connect the precision of mathematics with the complexities of the living world.