Computationally Tractable, Conceptually Plausible Classes of Link Matrices for the Inquery Inference Network
نویسنده
چکیده
This paper presents the PIC matrices, a computationally efficient subclass of link matrices that may be considered for the interpretation of query operators in the INQUERY inference network. The PIC class of matrices is formally defined; a algorithm, PIC-EVAL, is presented for the evaluation of members of this class; and a proof of the correctness of the algorithm is given. A further specialization of this class that can be evaluated even more efficiently is discussed. Finally, a generalization of the PIC class, for which the input probabilities may be viewed as weighted, is defined, and a simple modification of the PIC-EVAL algorithm that allows for the evaluation of matrices of this extended class is given.
منابع مشابه
Optimizing the Torque of Knee Movements of a Rehabilitation Robot
The aim of this study is to employ the novel Adaptive Network-based Fuzzy Inference System to optimize the torque applied on the knee link of a rehabilitation robot. Given the special conditions of stroke or spinal cord injury patients, devices with minimum error are required for performing the rehabilitation exercises. After examining the anthropometric data tables of human body, parameters su...
متن کاملTwo Methods for Wild Variational Inference
Variational inference provides a powerful tool for approximate probabilistic inference on complex, structured models. Typical variational inference methods, however, require to use inference networks with computationally tractable probability density functions. This largely limits the design and implementation of variational inference methods. We consider wild variational inference methods that...
متن کاملBridging the gap between theory and practice of approximate Bayesian inference
In computational cognitive science, many cognitive processes seem to be successfully modeled as Bayesian computations. Yet, many such Bayesian computations has been proven to be computationally intractable (NP-hard) for unconstrained input domains, even if only an approximate solution is sought. This computational complexity result seems to be in strong contrast with the ease and speed with whi...
متن کاملIndri at TREC 2004: Terabyte Track
This paper provides an overview of experiments carried out at the TREC 2004 Terabyte Track using the Indri search engine. Indri is an efficient, effective distributed search engine. Like INQUERY, it is based on the inference network framework and supports structured queries, but unlike INQUERY, it uses language modeling probabilities within the network which allows for added flexibility. We des...
متن کاملInquery and Trec-7 1.2 Inroute
ipated in only four of the tracks that were part of the TREC-7 workshop. We worked on ad-hoc retrieval, ltering, VLC, and the SDR track. This report covers the work done on each track successively. We start with a discussion of IR tools that were broadly applied in our work. Although UMass used a wide range of tools, from Unix shell scripts, to PC spreadsheets, three major tools were applied ac...
متن کامل