Dynamic Programming

By Richard Bellman. 1957

Citations in the corpus

• Reinforcement Learning in Healthcare: A Survey (2019, neutral): “ransition and reward functions, while model-free methods, also referred to as learning methods, learn an optimal policy simply based on the received observations and rewards. Dynamic programming (DP) [17] is a collection of modelbased techniques to compute an optimal policy given a complete descript”
• Hidden Markov processes (2002, unknown): “This problem is solved using dynamic programming [30] or by the well-known Viterbi algorithm [308], [124], [285].”
• An Introduction to Nonlinear Model Predictive Control (2002, unknown): “This follows from Bellman‘s Principle of Optimality [7], i.”
• Cyclic Equilibria in Markov Games (2005, unknown): “Value iteration (Bellman, 1957) has proven its worth in a variety of sequential-decisionmaking settings, most significantly single-agent environments (Puterman, 1994), team games, and two-player zero-sum games (Shapley, 1953).”
• Randomized Kinodynamic Planning (1999, unknown): “For problems that involve low degrees of freedom, classical dynamic programming ideas can be employed to yield numerical optimal control solutions (Bellman 1957; Bertsekas 1975; Larson and Casti 1982; LaValle 1998a).”
• An algorithm for optimal partitioning of data on an interval (2003, unknown): “ above problem by finding an exact global optimum, for any fitness function V that is additive in the sense of Eq. (4). There is a large (2N−1) but finite number of partitions in P∗. Dynamic programming [5] is an intelligent method of searching this space of all possible solutions. Our algorithm can be”
• Approximate Policy Iteration with a Policy Language Bias (2003, unknown): “Dynamic-programming approaches to finding optimal control policies in Markov decision processes (MDPs) [4, 14] using explicit (flat) state space representations break down when the state space becomes extremely large.”