1,721,000 research outputs found
Selecting Observation time in the monitoring and interpretation of Time-Varying data
No full textA lot of previous approaches to monitoring involved a continuous reading of the system parameters in order to recognize when anomalies in the behavior of the system under examination can trigger the diagnostic process. This paper deals with the application of Markov chain theory to the selection of observation time in the monitoring and diagnosis of time-varying systems. The goal of the present paper is to show how, by assuming a framework where the temporal behavior of the components of the system is modeled in a stochastic way, the continuous observation of critical parameters can be avoided; indeed, this kind of approach allows us to get a useful criterion for choosing observation time in domains where getting observations can be expensive. Observations are then requested only when the necessity for a diagnostic process becomes relevant and a focusing on the components that are more likely to be faulty can also be achieved
Utility-based approach to learning in a mixed case-based and model-based reasoning architecture
No full textCase-based reasoning (CBR) can be used as a form of "caching" solved problems to speedup later problem solving. Using "cached" cases brings additional costs with it due to retrieval time, case adaptation time and also storage space. Simply storing all cases will result in a situation in which retrieving and trying to adapt old cases will take more time (on average) than not caching at all. This means that caching must be applied selectively to build a case memory that is actually useful. This is a form of the utility problem [4, 2]. The approach taken here is to con struct a "cost model" of a system that can be used to predict the effect of changes to the system. In this paper we describe the utility problem associated with "caching" cases and the construction of a "cost model". We present experimental results that demonstrate that the model can be used to predict the effect of certain changes to the case memory
Deordering and numeric macro actions for plan repair
Full text linkThe paper faces the problem of plan repair in presence of numeric information, by providing a new method for the intelligent selection of numeric macro actions. The method relies on a generalization of deordering, extended with new conditions accounting for dependencies and threats implied by the numeric components. The deordering is used as a means to infer (hopefully) minimal ordering constraints then used to extract independent and informative macro actions. Each macro aims at compactly representing a sub-solution for the overall planning problem. To verify the feasibility of the approach, the paper reports experiments in various domains from the International Planning Competition. Results show (i) the competitiveness of the strategy in terms of coverage, time and quality of the resulting plans wrt current approaches, and (ii) the actual independence from the planner employe
Integrating Abductive Reasoning and Probabilistic Temporal Prediction in Diagnostic Problem Solving
No full textDiscusses an approach to diagnosis across different time instants, based on the decomposition of static and time-varying aspects; in particular the approach is based on the integration of abductive reasoning, used for interpreting observations at a given time point and probabilistic prediction concerning the temporal evolution of the components of the system to be diagnosed. The emphasis of the paper is on mechanisms for relating diagnostic hypotheses at different time instants. The authors show how the probability of the resulting histories can be computed by taking into account that partial diagnoses are produced by the abductive atemporal reasoner. The authors briefly discuss a prototype composed of two basic modules combined in a pipeline architecture, where the first module produces atemporal diagnoses that the second relates across different time point
Dynamic Case Memory Management
No full textThe present work describes some aspects related to the
utility/swamping problem in ADAPtER, a multimodal reasoning system
combining Case-Based Reasoning and Model-Based Reasoning
for diagnostic problem solving. A detailed set of experiments allowed
us to analyse the average behavior of the system with respect
to a given domain, in terms of performance of the whole system and
its components. Such experiments pointed out that the increasing of
the size of the case memory reduces the need for solving problem
from scratch, but is the main responsible for the arising of the utility
problem in ADAPtER. As a consequence, particular attention is
paid to the problem of dynamically maintaining under control the
growth of the case memory. We propose two learning strategies implementing
a dynamic approach to case memory management. Such
strategies allow the system to dynamically add or replace cases from
memory, in order to keep under control both case memory size and
content. Experimental testing of the above strategies suggests that
their adoption can greatly mitigate the over-sizing of the case memory
Utility-based approach to learning in a mixed case-based and model-based reasoning architecture
No full textCase-based reasoning (CBR) can be used as a form of "caching" solved problems to speedup later problem solving. Using "cached" cases brings additional costs with it due to retrieval time, case adaptation time and also storage space. Simply storing all cases will result in a situation in which retrieving and trying to adapt old cases will take more time (on average) than not caching at all. This means that caching must be applied selectively to build a case memory that is actually useful. This is a form of the utility problem [4, 2]. The approach taken here is to con struct a "cost model" of a system that can be used to predict the effect of changes to the system. In this paper we describe the utility problem associated with "caching" cases and the construction of a "cost model". We present experimental results that demonstrate that the model can be used to predict the effect of certain changes to the case memory
Automatic Case Base Management in a Multi-modal Reasoning System
No full textThe definition of suitable case base maintenance policies is widely recognized as a major success key of CBR systems; underestimating this issue may lead to systems that that do not perform adequately under performance dimensions, namely computation time, competence and quality of solutions. The goal of the present paper is to analyse an automatic case base management strategy in the context of multi-modal architectures combining CBR and Model-Based Reasoning. The strategy, called Learning by Failure with Forgetting (LFF ) is based on incremental learning of cases interleaved with off-line processes of case deletion, in order to control the content and the size of the case library. Results from an extensive experimental analysis in an industrial plant diagnosis domain is then reported, showing the usefulness of LFF with respect to the maintenance of suitable performance level for the target system
A comparative analysis of Horn models and Bayesian Networks for diagnosis
No full textThe aim of the paper is to formally relate logical Horn models and Bayesian Networks (BNs) in the framework of diagnostic reasoning. This is pursued by pointing out similarities between the two formalisms at the modeling level and by introducing into BNs a suitable notion of derivation. We also discuss modeling issues underlying the choice of Horn-based models vs BNs, by making explicit the “completion semantics” underlying a BN. This correspondence between “completed” Horn theories and BNs allows us to formally justify classical diagnostic schemata adopted for BNs
Combining Experiential Knowledge and Model-Based Reasoning for Diagnostic Problem Solving
No full text- …
