902 research outputs found
Towards Modular Compilation Using Higher-Order Effects
No full textCompilers transform a human readable source language into machine readable target language. Nanopass compilers simplify this approach by breaking up this transformation into small steps that are more understandable, maintainable, and extensible. We propose a semantics-driven variant of the nanopass compiler architecture exploring the use a effects and handlers to model the intermediate languages and the transformation passes, respectively. Our approach is fully typed and ensures that all cases in the compiler are covered. Additionally, by using an effect system we abstract over the control flow of the intermediate language making the compiler even more flexible. We apply this approach to a minimal compiler from a language with arithmetic and let-bound variables to a string of pretty printed X86 instructions. In the future, we hope to extend this work to compile a larger and more complicated language and we envision a formal verification framework from compilers written in this style
Prognostic Molecular Classification of Breast Cancer Based on Features Extracted from a Scale Space
No full textBreast cancer is one of the most prevalent cancers affecting females in the world. In recent years, many cancer researchers have been trying to determine molecular prognosis tools that predict cancer patient treatment response and/or chance of survival. In particular, the determination of gene expression signatures obtained by feature selection methods applied to large microarray datasets has shown potential. The main purpose of this study is to extend these gene signatures and molecular prognostic classifiers by investigating features constructed from a scale-space representation of the microarray data. Here, we construct a scale space by first mapping all genes to a one-dimensional functional space using protein family information. Next, we applied successive smoothing to the expression values resulting in one scale-space representation of the gene expression data from one sample. At the lowest scale, the scale space contains the original gene expression values, whereas at higher scales meta-features are formed, which are weighted sums of groups of genes. To test whether a scale-space representation is useful we performed feature selection and classification on a publicly available breast cancer expression dataset. We found that, instead of signatures consisting of single genes, meta-genes (i.e. groups of genes) that exist at higher scales were preferentially selected. We furthermore determined cross-validation errors using seven distinct classifiers (NMC, LDC, QDC, FISHERC, PARZENC, 3NNC, and LOGLC) and found that better performance is obtained using the scale-space representation than with the traditional representation of the gene expression data. As a result, we conclude that the scale-space analysis constitutes a potent way of selecting molecular signatures and is useful for prognostic classification.Pattern Recognition and BioinformaticsIntelligent SystemsElectrical Engineering, Mathematics and Computer Scienc
Identification and Elucidation of Expression Quantitative Trait Loci (eQTL) and their regulating mechanisms using Decodive Deep Learning
No full textMotivation: Identification and elucidation of eQTL has long been an active area of research. Finding cis-eQTL has been a manageable problem because of the limited number of candidates. Finding transeQTL has on the other hand been much more challenging because of the issue of multiple hypothesis testing. It has been suggested that additional information might alleviate this problem and although there has been some success using such methods no comprehensive data integration strategy has been developed. Approach: In order to comprehensively solve the issue of multiple hypothesis testing in the context of trans-eQTL discovery this research introduces MASSQTL: A comprehensive data integration method that makes use of a deep neural network (DNN) to prune the transeQTL candidate space to a desired size with the objective of finding more significantly associated trans-eQTL. Results: With MASSQTL many more trans-eQTL were found using a deep neural network filtering approach. The deep neural network outperformed other machine learning models showing that deep learning by use of complex hierarchical representations is able to model a diverse and sparse set of biological data. In addition to that the method provided new insight into the mechanisms underlying the regulatory architecture of gene expression.Computer SciencePattern Recognition and Bioinformatics groupElectrical Engineering, Mathematics and Computer Scienc
High performance parallelism pearls : multicore and many-core programming approaches / James Reinders, Jim Jeffers.
No full textElectronic reproduction. Palo Alto, Calif. : ebrary, 2014. Available via World Wide Web. Access may be limited to ebrary affiliated libraries.computer bookfair2016Includes indexes.Previously issued in print: 2014.xlv, 502 pages
Experiments in Bayesian Recommendation
No full textThe performance of collaborative filtering recommender systems can suffer when data is sparse, for example in distributed situations. In addition popular algorithms such as memory-based collaborative filtering are rather ad-hoc, making principled improvements difficult. In this paper we focus on a simple recommender based on naïve Bayesian techniques, and explore two different methods of modelling probabilities. We find that a Gaussian model for rating behaviour works well, and with the addition of a Gaussian-Gamma prior it maintains good performance even when data is sparse
Operating characteristics for the design and optimisation of classification systems
No full textIn statistical pattern recognition, problems involve distinguishing of various concepts or classes, based on the development of classifiers/discriminators. These exploit discriminatory information existing in measurements originating from objects. A trained classifier results in a partitioning in measurement space, providing some separation between the various classes. In the (typical) case of class overlap, this partitioning inherently results in a trade-off between the various possible classification errors that may occur. This partitioning can be modified to adjust these trade-offs. Given class abundances, a classifier can be evaluated at a given partitioning. However, variations in the abundances leads to an altered classifier performance. These fundamental aspects behind classifier design and evaluation can be studied within the framework of classifier operating characteristics, which is the topic of this dissertation. The contents consist of a number of published/accepted journal and conference papers, contextualised into a number of chapters representing various aspects of operating characteristic analysis. First the well-known two-class operating characteristic is considered, with two new analyses that are useful in certain circumstances. Next, the extension to the elusive multiclass case is considered, showing how standard 2-class operating characteristics analyses can be extended theoretically to the multiclass case. The challenge behind the multiclass extension is shown to be of a computational nature, with the calculation size increasing exponentially with the number of classes. The primary thesis contribution is then presented, consisting of a number of approaches and philosophies that can be used to overcome the computational challenges. Of primary importance is the finding that most practical problems are such that not all dimensions of the operating characteristic interact together significantly. Next it is shown how the operating characteristic approach can be used to design classifiers in ill-defined environments. In these problems some classes may be poorly represented, and the goal of the classifier design is to protect against these unforeseen conditions. Finally, it is shown that operating characteristics can be applied to a multi-stage classifier setup, allowing for a holistic design incorporating interactions between classes, and the classifier stages.Electrical Engineering, Mathematics and Computer Scienc
- …
