Learning methods based on dynamic programming (DP) are receiving increasing attention in arti cial intelligence. Researchers have argued that DP provides the appropriate basis for compiling planning results into reactive strategies for real-time control, as well as for learning such strategies when the system being controlled is incompletely known. We introduce an algorithm based on DP, which w...

In this paper, we show that the truly two-dimensional elastic image matching problem can be solved analytically using dynamic programming (DP) in polynomial time if the problem is formulated as a maximum a posteriori problem using Gaussian distributions for the likelihood and prior. After giving the derivation of the analytical DP matching algorithm, we evaluate its performance on handwritten c...

An alternative approach towards dynamic programming (DP) is presented: Recursions. A basic deterministic model and solution function are defined and the model is generalized to include stochastic processes, with the traditional stochastic DP model as a special case. Heuristic rules are included in the models simply as restrictions on decision spaces and a small slaughter pig marketing planning ...

Constraints over variable sequences are ubiquitous and many of their propagators have been inspired by dynamic programming (DP). We propose a conceptual framework for designing such propagators: pruning rules, in a functional notation, are refined upon the application of transformation operators to a DP-style formulation of a constraint; a representation of the (tuple) variable domains is picke...

We show how to significantly speed up MCMC sampling of DAG structures by using a powerful non-local proposal based on Koivisto’s dynamic programming (DP) algorithm (11; 10), which computes the exact marginal posterior edge probabilities by analytically summing over orders. Furthermore, we show how sampling in DAG space can avoid subtle biases that are introduced by approaches that work only wit...

In this paper an efficient segment selection method for corpusbased speech synthesis is presented. Traditional unit-selectors use dynamic programming (DP) to find in a fully connected segment network the most appropriate segment sequence based on targetand concatenation costs. Instead of performing a full DP search, the presented unit-selector applies fast transformations on binary valued segme...

An alternative approach towards dynamic programming (DP) is presented: Recursions. A basic deterministic model and solution function are defined and the model is generalized to include stochastic processes, with the traditional stochastic DP model as a special case. Heuristic rules are included in the models simply as restrictions on decision spaces and a small slaughter pig marketing planning ...

We discuss approximation algorithms based on Dynamic Programming techniques. Through Makespan and Knapsack, we see common ideas such as the dual view of parameters and bucketing and rounding values to a smaller distinct number of values, both of which can help facilitate the use of DP to approximate a solution. The Traveling Salesman Problem is also explored; the Euclidean version of TSP has a ...

This paper presents an automatic algorithm to extract the discrete phrase and accent commands with their timing parameters from the speech wave. This algorithm is based on dynamic programming (also called DP algorithm) and least square error methods. The DP algorithm is composed of two steps. First, it decides the timing of the phrase commands; then, it determines the suitable parameters of phr...

This study proposed a weighting function for a dynamic (DP) programming approach as the cost function to handle beat tracking. The function is utilized to adapt bandwidth of tempo deviation which reciprocal represents likelihood of relative beat position. The beat tracking surpass those scores of other algorithms for MAZ dataset, in which seven out of ten performance indexes outperforms in MIRE...