Results 1  10
of
10
Principles of Metareasoning
 Artificial Intelligence
, 1991
"... In this paper we outline a general approach to the study of metareasoning, not in the sense of explicating the semantics of explicitly specified metalevel control policies, but in the sense of providing a basis for selecting and justifying computational actions. This research contributes to a devel ..."
Abstract

Cited by 162 (10 self)
 Add to MetaCart
In this paper we outline a general approach to the study of metareasoning, not in the sense of explicating the semantics of explicitly specified metalevel control policies, but in the sense of providing a basis for selecting and justifying computational actions. This research contributes to a developing attack on the problem of resourcebounded rationality, by providing a means for analysing and generating optimal computational strategies. Because reasoning about a computation without doing it necessarily involves uncertainty as to its outcome, probability and decision theory will be our main tools. We develop a general formula for the utility of computations, this utility being derived directly from the ability of computations to affect an agent's external actions. We address some philosophical difficulties that arise in specifying this formula, given our assumption of limited rationality. We also describe a methodology for applying the theory to particular problemsolving systems, a...
A Bayesian Approach to Tackling Hard Computational Problems
 IN UAI
, 2001
"... We describe research and results centering on the construction and use of Bayesian models that can predict the run time of problem solvers. Our efforts are motivated by observations of high variance in the time required to solve instances for several challenging problems. The methods ..."
Abstract

Cited by 64 (10 self)
 Add to MetaCart
We describe research and results centering on the construction and use of Bayesian models that can predict the run time of problem solvers. Our efforts are motivated by observations of high variance in the time required to solve instances for several challenging problems. The methods
Principles and Applications of Continual Computation
 Artificial Intelligence
, 2001
"... Automated problem solving is viewed typically as the allocation of computational resources to solve one or more problems passed to a reasoning system. In response to each problem received, effort is applied in real time to generate a solution and problem solving ends when a solution is rendered. We ..."
Abstract

Cited by 40 (7 self)
 Add to MetaCart
Automated problem solving is viewed typically as the allocation of computational resources to solve one or more problems passed to a reasoning system. In response to each problem received, effort is applied in real time to generate a solution and problem solving ends when a solution is rendered. We examine continual computation, reasoning policies that capture a broader conception of problem by considering the proactive allocation of computational resources to potential future challenges. We explore policies for allocating idle time for several settings and present applications that highlight opportunities for harnessing continual computation in realworld tasks. 2001 Elsevier Science B.V. All rights reserved. Keywords: Bounded rationality; Decisiontheoretic control; Metareasoning; Deliberation; Compilation; Speculative execution; Value of computation 1.
How to Solve It Automatically: Selection Among ProblemSolving Methods
 Proceedings of the Fourth International Conference on Artificial Intelligence Planning Systems
, 1998
"... The choice of an appropriate problemsolving method, from available methods, is a crucial skill for experts in many areas. We describe a technique for the automatic selection among methods, which is based on a statistical analysis of their past performances. We formalize the statistical problem ..."
Abstract

Cited by 23 (0 self)
 Add to MetaCart
The choice of an appropriate problemsolving method, from available methods, is a crucial skill for experts in many areas. We describe a technique for the automatic selection among methods, which is based on a statistical analysis of their past performances. We formalize the statistical problem involved in selecting an efficient problemsolving method, derive a solution to this problem, and describe a methodselection algorithm. The algorithm not only chooses among available methods, but also decides when to abandon the chosen method, if it proves to take too much time. We give empirical results on the use of this technique in selecting among search engines in the PRODIGY planning system. 1 Introduction The choice of an appropriate problemsolving method is one of the main themes of Polya's famous book How to Solve It (Polya 1957). Polya showed that the selection of an effective approach to a problem is a crucial skill for a mathematician. Psychologists have accumulated m...
Solving TimeDependent Problems: A DecisionTheoretic Approach to Planning in Dynamic Environments
, 1991
"... Controlling a robot involves making decisions that modify its behavior. Making good decisions may require timeconsuming computation. Changes in the environment over time affect when this computation can be done (e.g., after obtaining the necessary information) , and when a result is useful (e.g., b ..."
Abstract

Cited by 15 (1 self)
 Add to MetaCart
Controlling a robot involves making decisions that modify its behavior. Making good decisions may require timeconsuming computation. Changes in the environment over time affect when this computation can be done (e.g., after obtaining the necessary information) , and when a result is useful (e.g., before some event occurs). This sensitivity to when computation is performed and when decisions are made is what makes these problems "timedependent." A controller with more than one decision to make must trade off computation time, based on the expected effect on the system's behavior. We call the resulting metalevel scheduling problem a "deliberationscheduling" problem. We have
Reasoning, Metareasoning, and Mathematical Truth: Studies of Theorem Proving under Limited Resources
 In Proceedings of UAI95
, 1995
"... In earlier work, we introduced flexible inference and decisiontheoretic metareasoning to address the intractability of normative inference. Here, rather than pursuing the task of computing beliefs and actions with decision models composed of distinctions about uncertain events, we examine methods f ..."
Abstract

Cited by 11 (3 self)
 Add to MetaCart
In earlier work, we introduced flexible inference and decisiontheoretic metareasoning to address the intractability of normative inference. Here, rather than pursuing the task of computing beliefs and actions with decision models composed of distinctions about uncertain events, we examine methods for inferring beliefs about mathematical truth before an automated theorem prover completes a proof. We employ a Bayesian analysis to update belief in truth, given theoremproving progress, and show how decisiontheoretic methods can be used to determine the value of continuing to deliberate versus taking immediate action in timecritical situations. 1 INTRODUCTION Theorem proving is frequently perceived as a logical, deterministic endeavor. However, uncertainty often plays a critical role in theorem proving and other mathematical pursuits. The mathematician George Polya emphasized the importance of plausible reasoning for guiding the intuitions and effort of mathematicians. In particular, h...
Automatic Evaluation and Selection of ProblemSolving Methods: Theory and Experiments
 Journal of Experimental and Theoretical Artificial Intelligence
"... The choice of the right problemsolving method, from available methods, is a crucial skill for experts in many areas. We present a technique for automatic selection among methods based on analysis of their past performances. We formalize the statistical problem involved in choosing an e#cient met ..."
Abstract

Cited by 5 (0 self)
 Add to MetaCart
The choice of the right problemsolving method, from available methods, is a crucial skill for experts in many areas. We present a technique for automatic selection among methods based on analysis of their past performances. We formalize the statistical problem involved in choosing an e#cient method, derive a solution to this problem, and describe a selection algorithm. The algorithm not only chooses among available methods, but also decides when to abandon the chosen method if it takes too much time. We then extend the basic statistical technique to account for problem sizes and similarity among problems.
Thinking Ahead: Continual Computation Policies for Allocating Idle and RealTime Resources to Solve Future Challenges
, 1999
"... Research on continual computation centers on developing precomputation policies that can effectively harness available resources to solve future challenges. We focus on integrating a consideration of offline and realtime resources in continual computation. We review precomputation policies for flex ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
Research on continual computation centers on developing precomputation policies that can effectively harness available resources to solve future challenges. We focus on integrating a consideration of offline and realtime resources in continual computation. We review precomputation policies for flexible procedures and present strategies that account for the expected future realtime refinement of a result following precomputation. Finally, we address policies that consider the tradeoff between the efficiency of solving current and potential future challenges.
Principles of Efficient Inference
, 2001
"... The goal of my research is to uncover fundamental principles for the construction of realtime AI systems that employ declarative knowledge representations and general reasoning engines. There are many advantages to such an architecture: The same knowledge can be used for multiple tasks, such as dia ..."
Abstract
 Add to MetaCart
The goal of my research is to uncover fundamental principles for the construction of realtime AI systems that employ declarative knowledge representations and general reasoning engines. There are many advantages to such an architecture: The same knowledge can be used for multiple tasks, such as diagnosis, prediction, control, and explanation. General knowledge can be applied to novel problems and environments. New, improved reasoning engines can be used without reengineering the entire system.
kautz~cs.washington.edu selman~cs.cornell.edu
"... dmax~microsoft.com We describe research and results centering on the construction and use of Bayesian models that can predict the run time of problem solvers. Our efforts are motivated by observations of high variance in the time required to solve instances for several challenging problems. The meth ..."
Abstract
 Add to MetaCart
dmax~microsoft.com We describe research and results centering on the construction and use of Bayesian models that can predict the run time of problem solvers. Our efforts are motivated by observations of high variance in the time required to solve instances for several challenging problems. The methods have application to the decisiontheoretic control of hard search and reasoning algorithms. We illustrate the approach with a focus on the task of predicting run time for general and domainspecific solvers on a hard class of structured constraint satisfaction problems. We review the use of learned models to predict the ultimate length of a trial, based on observing the behavior of the search algorithm during an early phase of a problem session. Finally, we discuss how we can employ the models to inform dynamic runtime decisions. 1