Open Access. Powered by Scholars. Published by Universities.®
Physical Sciences and Mathematics Commons™
Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 3 of 3
Full-Text Articles in Physical Sciences and Mathematics
Maintaining High Performance Across All Problem Sizes And Parallel Scales Using Microkernel-Based Linear Algebra, Md Rakib Hasan
Maintaining High Performance Across All Problem Sizes And Parallel Scales Using Microkernel-Based Linear Algebra, Md Rakib Hasan
LSU Doctoral Dissertations
Linear algebra underlies a large proportion of computational problems. With the continuous increase of scale on modern hardware, performance of small sized linear algebra has become increasingly important. To overcome the shortcomings of conventional approaches, we employ a new approach using a microkernel framework provided by ATLAS to improve the performance of a few linear algebra routines for all problem sizes. Our initial research consists of improving the performance of parallel LU factorization in ATLAS for which we were able to achieve up to 2.07x and 2.66x speedup for small problems, up to 91% and 87% of theoretical peak performance …
The Impact Of Overfitting And Overgeneralization On The Classification Accuracy In Data Mining, Huy Nguyen Anh Pham
The Impact Of Overfitting And Overgeneralization On The Classification Accuracy In Data Mining, Huy Nguyen Anh Pham
LSU Doctoral Dissertations
Current classification approaches usually do not try to achieve a balance between fitting and generalization when they infer models from training data. Such approaches ignore the possibility of different penalty costs for the false-positive, false-negative, and unclassifiable types. Thus, their performances may not be optimal or may even be coincidental. This dissertation analyzes the above issues in depth. It also proposes two new approaches called the Homogeneity-Based Algorithm (HBA) and the Convexity-Based Algorithm (CBA) to address these issues. These new approaches aim at optimally balancing the data fitting and generalization behaviors of models when some traditional classification approaches are used. …
Model-Driven Search-Based Loop Fusion Optimization For Handwritten Code, Pamela Bhattacharya
Model-Driven Search-Based Loop Fusion Optimization For Handwritten Code, Pamela Bhattacharya
LSU Master's Theses
The Tensor Contraction Engine (TCE) is a compiler that translates high-level, mathematical tensor contraction expressions into efficient, parallel Fortran code. A pair of optimizations in the TCE, the fusion and tiling optimizations, have proven successful for minimizing disk-to-memory traffic for dense tensor computations. While other optimizations are specific to tensor contraction expressions, these two model-driven search-based optimization algorithms could also be useful for optimizing handwritten dense array computations to minimize disk to memory traffic. In this thesis, we show how to apply the loop fusion algorithm to handwritten code in a procedural language. While in the TCE the loop fusion …