Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 46 for “"Structured Sparsity"”.
-
Structured Sparsity Learning for Coevolution-Based Protein Contact Prediction
… type information, to characterize the desired sparsity patterns of contact prediction in a biologically meaningful way. It then uses multiple structured sparsity regularization models, including group LASSO and group dispersive sparsity, to enforce such sparsity patterns. This method benefits …
-
Learning to Measure: Distance Metric Learning with Structured Sparsity
… and put forth solutions which make use of structured sparsity. The contributions of this dissertation may be broadly divided into two portions. In the first portion (chapter 2) we begin with a classic approach in distance metric learning and address a scenario where distance metric learning …
-
Sparse Coding with Structured Sparsity Priors and Multilayer Architecture for Image Classification
… are expected to combine the best of both sparsity regularizations and deep architectures. To date, however, endeavors to marry the two techniques have not achieved significant improvements over their individual counterparts. In this thesis, we first briefly review multiple structured …
-
Structured Sparsity-Aware Hardware-Software Co-Design for Deep Neural Network Acceleration
… and energy efficiency. Using various types of sparsity in DNNs has been proposed recently to reduce compute and memory requirements, but finding the proper target sparsity to meet both HW and SW requirements is still an active area of research. In this thesis, we develop HW-SW co-design methods …
-
Binary Sparse Modeling for Multi-Anchor Direct Localization using Channel State Information
… modeling, this thesis explores the concept of structured sparsity in the received signals across anchors. It introduces a binary approximation method to address the challenges of structured sparsity, enhancing the precision and reliability of support recovery solutions. This approach …
-
SPARSITY-BASED REPRESENTATION WITH LOW-RANK INTERFERENCE: ALGORITHMS AND APPLICATIONS
… interference while simultaneously promoting sparsity-based representation of multiple correlated signals. The proposed framework provides a new and efficient approach for the representation of multiple measurements where the underlying signals exhibit a structured sparsity representation over …
-
Structured Dictionary Learning and its applications in Neural Recording
… for classification purpose. However, current sparsity‐based approaches for neural recording are limited due to several critical drawbacks: (i) the lack of an efficient data‐driven representation to fully capture the characteristics of specific neural signal; (ii) most existing methods do not …
-
Bi-convex model based reconstruction methods for magnetic resonance elastography
… model-based reconstruction techniques with structured sparsity for faster MRE acquisition and robust elastogram reconstruction. A new optimization technique for elastogram reconstruction is proposed using the bi-convexity of the elastodynamic model and the alternating direction method of …
-
Recovery of Compressive Sensed Images With Piecewise Autoregressive Modeling
… offer a powerful mechanism to characterize structured sparsity of natural images. This mechanism greatly restricts the CS solution space. Simulation results show the preeminent effect of our algorithms in the recovery of wide range of natural images. In average our best algorithm improves …
-
Interpretable Learning for Hospital Readmission Prediction from Healthcare Data
… algorithm. In the first approach, a structured sparsity regularization based model is applied. In the second approach, to improve the interpretability, a novel tree-structured sparsity-inducing regularization norm is proposed. Furthermore, a quantitative evaluation metric to assess …
-
TOWARDS IMPROVING GENERALIZATION OF DEEP NEURAL NETWORKS: DATA, PARAMETER OPTIMIZATION, AND DEPLOYMENT
… by combining zeroth-order optimization with structured sparsity. From the deployment perspective, I introduce SeedLoRA, a parameter-efficient model fusion method that merges adapters trained with different random seeds directly in weight space. This strategy uses the diversity of independent …
-
MODEL ADAPTATION FOR EDGE AI
… pruning framework that employs a hybrid structured sparsity pattern. Thirdly, while low-precision integer formats are common for reduced precision arithmetic, these formats are inadequate for emerging workloads like transformers. We explore minifloats, reduced-precision floating-point …
-
Methods for Optimizing BERT Model on Edge Devices - Accelerating Biomedical NLP with Pruned and Quantized BERT Models
… pruned with a three-stage, Fisher-guided structured pruning workflow at three levels of sparsity. Subsequently, in the second stage, dynamic 8-bit integers quantization using ONNX Runtime was applied to standard BERT, BioBERT, and DistilBERT. The third stage involved combining both pruning …
-
High-Dimensional Covariate-Dependent Gaussian Graphical Models
… This model incorporates covariate effects and structured sparsity (through colorings) to model dynamic conditional dependencies. We perform model estimation using penalized composite likelihood, employing coordinate descent and Broyden’s method for optimization under different scenarios. We …
-
A Reconfigurable Architecture for Dense and Sparse Matrix Multiplication in Processing-in-Memory
… lack the capability to efficiently handle matrix sparsity, limiting their ability to fully exploit the redundancy present in modern workloads. This work proposes a reconfigurable hardware architecture for efficient matrix-vector multiplication (MVM) within a PIM framework, designed to natively …
-
Optimization Strategies in Proton Treatment Planning Considering Biological Effects
… (LET) into the selection process. A group-sparsity formulation with a proximal semi-stochastic gradient descent algorithm is developed to jointly optimize dose and LET distributions, enabling beam selection that enhances tumor effectiveness while reducing LET in critical structures. The …
-
Novel Machine Learning Methods For Modeling Time-To-Event Data
… intra-event correlations in survival data using structured sparsity-based regularizers, (ii) learning novel representations for survival data by inferring inter-event and intra-event correlations, (iii) extending linear regression-based methods to learn predictive models from right censored data …
-
Structured sparse models for classification
… classification of high dimensional data using structured sparsity. Sparse models, where data is assumed to be well represented as a linear combination of a few elements from a dictionary, have gained considerable attention in recent years, and its use has led to state-of-the-art results in many …
Page 1 of 3