Paradoxical results on object-recognition delayed nonmatching-to-sample (DNMS) tasks have been found in monkeys and rats that receive either partial, ischemia-induced hippocampal lesions or complete hippocampal ablation. Ischemia results in severe DNMS impairments, which have been attributed to circumscribed CA1 cell loss. However, ablation studies indicate that the hippocampus plays only a minimal role in the performance of the DNMS task. Two hypotheses have been proposed to account for these discrepant findings (Bachevalier & Mishkin, 1989). First, the "hippocampal interference" hypothesis posits that following ischemia, the partially damaged hippocampus may disrupt activity in extrahippocampal structures that are important for object-recognition memory. Second, previously undetected ischemia-induced extrahippocampal damage may be responsible for the DNMS impairments attributed to CA1 cell loss. To test the "hippocampal interference" hypothesis, the effect of partial NMDAinduced lesions of the dorsal hippocampus were investigated on DNMS performance in rats. These lesions damaged much of the same area, the CA1, as did ischemia; but did so without depriving the entire forebrain of oxygen, thereby reducing the possibility of extrahippocampal damage. In Experiment 1, rats were trained on the DNMS task prior to receiving an NMDA-lesion. Postoperatively, these rats reacquired the nonmatching rule at a rate equivalent to controls and were unimpaired in performance at delays up to 300 s. In Experiment 2, naive rats were given NMDA-lesions and then trained on DNMS. These rats acquired the DNMS rule at a rate equivalent to controls and performed normally at delays up to 300 s. These findings suggest that interference from a partially damaged hippocampus cannot account for the ischemia-induced DNMS impairments and that they are more likely produced by extrahippocampal neuropathology. In Experiment 3, rats from the previous study were tested on the Morris water-maze. Compared to sham-lesioned animals, rats with partial lesions of the dorsal hippocampus were impaired in the acquisition of the water-maze task. Thus, subtotal NMDA-lesions of the hippocampus impaired spatial memory while leaving nonspatial memory intact. Mumby et al. (1992b) suggested that the ischemia-induced extrahippocampal damage underlying the DNMS deficits is mediated or produced by the postischemic hippocampus. To test this idea, preoperatively trained rats in Experiment 4 were subject to cerebral ischemia followed within 1hr by hippocampal aspiration lesions. It was hypothesized that ablation soon after ischemia would block the damage putatively produced by the postischemic hippocampus and thereby prevent the development of postoperative DNMS deficits. Unlike "ischemia-only" rats, the rats with the combined lesion were able to reacquire the nonmatching rule at a normal rate and performed normally at delays up to 300 s. Thus, hippocampectomy soon after ischemia eliminated the pathogenic process that lead to ischemia-induced DNMS deficits. Experiment 5 investigated the role of ischemiainduced CA1 cell death as a factor in the production of extrahippocampal neuropathology. Naive rats were given NMDA-lesions of the dorsal hippocampus followed 3 weeks later by cerebral ischemia. If the ischemia-induced CA1 neurotoxicity is responsible for producing extrahippocampal damage then preischemic ablation should attenuate this process and prevent the development of DNMS impairments. This did not occur: Rats with the combined lesion were as impaired as the "ischemia-only" rats in the acquisition of the DNMS task. This suggests that the ischemia-induced pathogenic processes that result in extrahippocampal neuropathology comprise more than CA1 neurotoxicity. The findings presented in this thesis are consistent with the idea that ischemiainduced DNMS deficits in rats are the result of extrahippocampal damage mediated or produced by the postischemic hippocampus. The discussion focuses on three main points: 1) How might the post-ischemic hippocampus be involved in the production of extrahippocampal neuropathology? 2) In what brain region(s) might this damage be occurring? 3) What anatomical, molecular, or functional neuropathology might ischemia produce in extrahippocampal brain regions? The results are also discussed in terms of a specialized role for the hippocampus in mnemonic functions and the recently emphasized importance of the rhinal cortex in object-recognition memory. / Arts, Faculty of / Psychology, Department of / Graduate

Machine Learning (ML) is a transformative technology at the forefront of many modern research endeavours. The technology is generating a tremendous amount of attention from researchers and practitioners, providing new approaches to solving complex classification and regression tasks. While concepts such as Deep Learning have existed for many years, the computational power for realising the utility of these algorithms in real-world applications has only recently become available. This dissertation investigated the efficacy of a novel, general method for deploying ML in a variety of complex tasks, where best feature selection, data-set labelling, model definition and training processes were determined automatically. Models were developed in an iterative fashion, evaluated using both training and validation data sets. The proposed method was evaluated using three distinct case studies, describing complex classification tasks often requiring significant input from human experts. The results achieved demonstrate that the proposed method compares with, and often outperforms, less general, comparable methods designed specifically for each task. Feature selection, data-set annotation, model design and training processes were optimised by the method, where less complex, comparatively accurate classifiers with lower dependency on computational power and human expert intervention were produced. In chapter 4, the proposed method demonstrated improved efficacy over comparable systems, automatically identifying and classifying complex application protocols traversing IP networks. In chapter 5, the proposed method was able to discriminate between normal and anomalous traffic, maintaining accuracy in excess of 99%, while reducing false alarms to a mere 0.08%. Finally, in chapter 6, the proposed method discovered more optimal classifiers than those implemented by comparable methods, with classification scores rivalling those achieved by state-of-the-art systems. The findings of this research concluded that developing a fully automated, general method, exhibiting efficacy in a wide variety of complex classification tasks with minimal expert intervention, was possible. The method and various artefacts produced in each case study of this dissertation are thus significant contributions to the field of ML.

No description available.

No description available.

No description available.

The problem of interest as indicated is to develop a general purpose technique that is a combination of the structural approach, and an extension of the Finite Inductive Sequence (FI) technique. FI technology is pre-algebra, and deals with patterns for which an alphabet can be formulated.

Neural networks have frequently been found to give accurate solutions to hard classification problems. However neural networks do not make explained classifications because the class boundaries are implicitly defined by the network weights, and these weights do not lend themselves to simple analysis. Explanation is desirable because it gives problem insight both to the designer and to the user of the classifier. Many methods have been suggested for explaining the classification given by a neural network, but they all suffer from one or more of the following disadvantages: a lack of equivalence between the network and the explanation; the absence of a probability framework required to express the uncertainty present in the data; a restriction to problems with binary or coarsely discretised features; reliance on axis-aligned rules, which are intrinsically poor at describing the boundaries generated by neural networks. The structure of the solution presented in this thesis rests on the following steps: Train a standard neural network to estimate the class conditional probabilities. Bayes’ rule then defines the optimal class boundaries. Obtain an explicit representation of these class boundaries using a piece-wise linearisation technique. Note that the class boundaries are otherwise only implicitly defined by the network weights. Obtain a safe but possibly partial description of this explicit representation using rules based upon the city-block distance to a prototype pattern. The methods required to achieve the last two represent novel work which seeks to explain the answers given by a proven neural network solution to the classification problem.

No description available.

Thesis (Ph. D.)--Oregon State University, 2009. / Printout. Includes bibliographical references (leaves 132-137). Also available on the World Wide Web.

Thesis (Ph. D.)--University of Texas at Austin, 1999. / Vita. Includes bibliographical references (leaves 107-114). Available also in a digital version from Dissertation Abstracts.