Showing papers on "Admissible heuristic published in 2010"

Common Misconceptions Concerning Heuristic Search

Abstract: This paper examines the following statements about heuristic search, which are commonly held to be true: More accurate heuristics result in fewer node expansions by A* and IDA*. A* does fewer node expansions than any other equally informed algorithm that finds optimal solutions. Any admissible heuristic can be turned into a consistent heuristic by a simple technique called pathmax. In search spaces whose operators all have the same cost A* with the heuristic function h(s)=0 for all states, s, is the same as breadth-first search. Bidirectional A* stops when the forward and backward search frontiers meet. The paper demonstrates that all these statements are false and provides alternative statements that are true.

Searching without a heuristic: efficient use of abstraction

Abstract: In problem domains where an informative heuristic evaluation function is not known or not easily computed, abstraction can be used to derive admissible heuristic values. Optimal path lengths in the abstracted problem are consistent heuristic estimates for the original problem. Pattern databases are the traditional method of creating such heuristics, but they exhaustively compute costs for all abstract states and are thus usually appropriate only when all instances share the same single goal state. Hierarchical heuristic search algorithms address these shortcomings by searching for paths in the abstract space on an as-needed basis. However, existing hierarchical algorithms search less efficiently than pattern database constructors: abstract nodes may be expanded many times during the course of a base-level search. We present a novel hierarchical heuristic search algorithm, called Switchback, that uses an alternating direction of search to avoid abstract node re-expansions. This algorithm is simple to implement and demonstrates superior performance to existing hierarchical heuristic search algorithms on several standard benchmarks.

Finding Acceptable Solutions Faster Using Inadmissible Information

Abstract: Bounded suboptimal search algorithms attempt to find a solution quickly while guaranteeing that the cost does not exceed optimal by more than a desired factor. These algorithms generally use a single admissible heuristic both for guidance and guaranteeing solution quality. We present a new approach to bounded suboptimal search that separates these roles, consulting multiple sources of potentially inadmissible information to determine search order and using admissible information to guarantee quality. An empirical evaluation across six benchmark domains shows the new approach has better overall performance.

Monte-Carlo Tree Search for a reentrant scheduling problem

Abstract: This paper reports application results of Monte-Carlo Tree Search (MCTS) for a practical problem. In this paper, a reentrant scheduling problem is considered as a practical problem which has been addressed by our previous works. MCTS introduced by Coulom is a best-first search where the pseudorandom simulations guide the solution of problem. Recent improvements on MCTS have produced strong computer Go program, which has a large search space, and the success is a hot topic for selecting the best move. So far, most of reports about MCTS have been on two-player game, and MCTS has been applied rarely in one-player perfect-information games. MCTS does not need an admissible heuristic, so the application of MCTS for one-player games might be an interesting alternative. Additionally, one-player games like puzzles are determinately operated only by one player's decision, so sequences of changes in state are describable as a network diagram of interdependence of operations. Therefore if MCTS for one-player games is considered as a meta-heuristic algorithm, we can use this algorithm for not only many practical problems, but also combinatorial optimization problems. Especially as MCTS does not fully depend on evaluation function, so the solutions based on MCTS remain effective if objective function of problem is modified. This paper firstly investigated on the application of Single Player MCTS (SP-MCTS) introduced by Schadd et al. Next this paper showed the effectiveness of new simulation strategies on SP-MCTS by numerical experiments. Based on the results, this paper discussed the application potentiality of SP-MCTS for a practical reentrant scheduling problem.

A New Approach of Iterative Deepening Bi- Directional Heuristic Front-to-Front Algorithm (IDBHFFA)

Abstract: Artificial Intelligence (AI) is a subject that studies techniques for making computers exhibit intelligent behavior. S earching still remains one of the problem in AI. Bi -directional search is performed by searching simultaneously in forward direction from the initial node and in backward direction from the goal node. Bi-directional heuristic search algorithms need less time and space than their unidirectional versions. Bi -directional Heuristic Front to Front Algorithm (BHFFA) is one of the Bi - directional heuristic search algorithm. However, it has some disadvantages. It needs to store many unnecessary nodes prior to termination. Moreover, in large problem spaces the computational overhead for the selection of the next node to be expanded increases significantly. This paper presents a modification to the BHFFA called Iterative Deepening Bi - directional Heuristic Front-to-Front Algorithm (IDBHFFA) that has been analyzed and implemented using the 8-puzzle problem. The proposed algorithm performs BHFFA in a number of iterations. For each iteration, two thresholds are maintained, one for each search frontier. In each iteration, some non-promising nodes on a particular search frontier having cost estimates greater than the corresponding threshold value are pruned. The process continues with successive iterations with the thresholds increasing with each iteration. The proposed algorithm will return optimal solutions with an admissible heuristic function. It can minimize the computational time and memory space requirement of BHFFA considerably.

Top-Down Search with Bottom-Up Evidence for Recognizing Handwritten Mathematical Expressions

Abstract: In handwritten mathematical expressions (ME), understanding the general structure of an ME is often easier than resolving local ambiguities. For instance, identifying a key operator in terms of its spatial relationship with its subordinates is relatively easier than resolving the ambiguities of single symbol identity and local spatial relationships. In addition, decisions related to key operators often occur close to the top (root) of the parse tree, while local decisions take place at the bottom of it. Based on these observations, we propose an incremental search framework in which a parse tree is expanded by tentatively selecting the key operators of an expression. The goodness of the selection is defined by the likelihood of key symbol, the goodness of the sub expressions, and their spatial relationships. In this framework, ambiguous local parts are processed after tentative decisions have been made at the global level. To handle explosiveness of key operator selection, an admissible heuristic function is defined based on the direct relationship of the key operator with the symbols at the bottom level. An experimental evaluation shows that our system is promising. Using it a robust interpretation can be made by utilizing global information and an interpretation can be reached quickly by the admissible heuristic function.

Observation planning with on-line algorithms and GPU heuristic computation

Abstract: When making an useful description of its environment, a robot has to identify both the free space and the objects location. SLAM algorithms are used for computing the free space map and an image processing algorithm is used in order to identify the objects. Those algorithms are time consuming (the time to go to the observation location and the image processing time) and are not perfect (their outcomes are stochastic). Furthermore, the agent may have multiple target to identify at the same time, and so has to build a policy for identication. We propose a Markov Decision Process (MDP)-based approach to compute those policies. Since in our application, the policy has to be computed on-line, in a limited time, optimal algorithms are too slow for those on-line purposes with a large state space. We show how on-line approaches oer solutions to the observation planning problem, and how to eciently use those algorithms by computing an accurate admissible heuristic on a GPU.

Continual On-Line Planning

Abstract: My research represents an approach to integrating planning and execution in time-sensitive environments. The primary focus is on a problem called continual on-line planning. New goals arrive stochastically during execution, the agent issues actions for execution one at a time, and the environment is otherwise deterministic. My dissertation will address this setting in three stages: optimizing total goal achievement time, handling on-line goal arrival during planning or execution, and adapting to changes in state also during planning or execution. My current approach to this problem is based on incremental heuristic search. The two central issues are the decision of which partial plans to elaborate during search and the decision of when to issue an action for execution. I have proposed an extension of Russell and Wefald's decision-theoretic A* algorithm that is not limited by assumptions of an admissible heuristic like DTA*. This algorithm, Decision Theoretic On-line Continual Search (DTOCS), handles the complexities of the on-line setting by balancing deliberative planning and real-time response.

Finding Acceptable Solutions Faster Using Inadmissible Information Computer Science Technical Report 10-01

Abstract: Bounded suboptimal search algorithms attempt to find a solution quickly while guaranteeing that its cost does not exceed optimal by more than a desired factor. Typically these algorithms use a single admissible heuristic evaluation function for both guiding search and bounding solution quality. In this paper, we present a new approach to bounded suboptimal search that separates these roles, consulting inadmissible information to determine search order and using admissible information to guarantee quality. Unlike previous proposals, it explicitly estimates expected solution cost and search distance in an attempt to reach a solution within the suboptimality bound as quickly as possible. We show how to construct these estimates during search using information that is readily available yet often overlooked. In an empirical evaluation across six diverse benchmark domains, the new techniques have better overall performance than previous approaches, including weighted A* and optimistic search.

Landmarks, Critical Paths and Abstractions: What's the Difference Anyway?

Abstract: Current heuristic estimators for classical domain-independent planning are usually based on one of four ideas: delete relaxation, abstraction, critical paths, and, most recently, landmarks. Previously, these different ideas for deriving heuristic functions were largely unconnected. In my talk, I will show that these heuristics are in fact very closely related. Moreover, I will introduce a new admissible heuristic called the landmark cut heuristic which exploits this relationship. In our experiments, the landmark cut heuristic provides better estimates than other current admissible planning heuristics, especially on large problem instances.

Evaluating Simulation Heuristics in Monte-Carlo Tree Search and its Application to a Production Scheduling

Abstract: This paper reports simulation heuristics of Monte‐Carlo Tree Search (MCTS) and shows an application example. MCTS introduced by Coulom is a best‐first search where pseudorandom simulations guide the solution of problem. Recent improvements on MCTS have produced strong computer Go program, which has a large search space, and the success is a hot topic for selecting the best move. So far, most of reports about MCTS have been on two‐player games, and MCTS has been used rarely for one‐player perfect‐information games. MCTS does not need admissible heuristic, so the application of MCTS for one‐player games might be an interesting alternative. Additionally, one‐player games like puzzles are determinately operated only by one player’s decision, so the sequences of changes in state are describable as a network diagram with interdependence between operations. If MCTS for one‐player games is available as a meta‐heuristic algorithm, we can use this algorithm for not only combinatorial optimization problems, but also many practical problems. Especially, as MCTS does not fully depend on evaluation function, so the solutions based on MCTS remain effective if objective function is modified. This paper firstly investigates on the application of Single Player MCTS (SP‐MCTS) introduced by Schadd et al. to a puzzle game called Bubble Breaker. Next this paper shows the effectiveness of new simulation strategies of SP‐MCTS, and considers the differences between each parameter. Based on the results, this paper discusses the application potentiality of SP‐MCTS for a scheduling problem.