Improving Deep Representations by Incorporating Domain Knowledge and Modularization for Synthetic Aperture Radar and Physiological Data
- Project Research
- 1-5 Chapters
- Abstract : Available
- Table of Content: Available
- Reference Style: APA
- Recommended for : Student Researchers
- NGN 3000
Abstract
Machine Learning (ML) using Artificial Neural Networks (ANNs), referred to as Deep Learning (DL), is a very popular and powerful method of statistical inference. A primary advantage of deep-learning has been the automatic learning of informative features (that encodes the data referred to as deep-representations henceforth) based on gradient-descent optimization of an objective function. While DL is applicable to problem domains where hand-crafted features are not readily available, its performance is critically dependent on other factors like dataset size and model architecture. Despite recent advances in the field, the question of how to modify the DL framework to incorporate domain knowledge or to disentangle factors of variation warrants more research. Until recently, most popular works in the DL literature have primarily employed inductive-bias of architectures (e.g., translational invariance in convolutional neural-nets) and relied on the availability of large labeled datasets for improved representation learning. Unfortunately, curating such large datasets is costly and not practical for many application areas. In this dissertation, we study methods to improve learned representations by incorporating domain knowledge into the learning process and through disentangling factors of variation. First, we present a sparse-modeling based data augmentation method for tomographic images and use it to incorporate domain knowledge of Synthetic Aperture ii Radar (SAR) target phenomenology into deep representations. We validate the improvements in learned representations by using them for a benchmark classification problem of Automatic Target Recognition (ATR) where we establish new state-of-theart on subsampled datasets. Second, we propose a DL-based hierarchical modeling strategy for a physiological signal generation process which in turn can be used for data augmentation. Based on the physiology of cardiovascular system function, we propose a modularized hierarchical generative model and then impose explicit regularizing constraints on each module using multi-objective loss functions. This generative model, called CardioGen, is evaluated by its ability to augment real data while training DL based models. The proposed approach showed performance improvements. Third, we propose a hierarchical deep-generative model for SAR imagery that jointly captures the underlying structure of multiple resolutions of SAR images. We utilize this model, called MrSARP, to super-resolve lower resolution magnitude images to a higher resolution. We evaluate the model’s performance using the three standard error metrics used for evaluating super-resolution performance on simulated data. Fourth, we propose a framework for learning a sufficient statistic of the data for a given downstream inference task. We design and train a DL model that encodes the Photoplethysmography (PPG) signal to a sufficient statistic and decodes it back to a task-specific PPG-like signal assuming it will be used for a fixed RR-tachogram prediction task. Compression and privacy-preservation can be a useful side-benefit of having such a downstream task. The learned deep representations of PPG data are validated using tachogram prediction error as well as its performance on the sub-task of stress estimation. i
Related Project Materials
AN EXAMINATION OF WESTERN MEDIA AND ITS INFLUENCE ON THE FASHION CHOICES OF NIGERIAN LADIES
EXCERPT FROM THE STUDY
"Western culture is an extremely broad term that refers to the social norms, belief systems, traditions, cust...
THE ASSESSMENT OF HUMAN RESOURCE DEVELOPMENT IN ENHANCING ORGANIZATIONAL PRODUCTIVITY
ABSTRACT
This project is on ‘’the assessment of Human Resource Development in enhancin...
THE ROLE OF ADVERTISING IN MOTIVATING CONSUMER BRAND PREFERENCE FOR BEVERAGES
EXCERPT FROM THE STUDY
Advertising as a promotional or communicational tool is an aid to trade for driving sales o...
THE IMPACT OF BUDGETING AND BUDGETARY CONTROL IN CONSTRUCTION PROJECT DELIVERY IN NIGERIA
Abstract
The study intends to investigate the impact of budgeting and budgetary control in construction project delivery...
THE RELATIONSHIP BETWEEN EFFECTIVE CLAIMS MANAGEMENT AND THE GROWTH OF NIGERIAN INSURANCE INDUSTRY
EXCERPT FROM THE STUDY
The existence of insurance markets facilitates economic ...
THE IMPACT OF TAX TREATIES ON INTERNATIONAL TAX PLANNING
Abstract: THE IMPACT OF TAX TREATIES ON INTERNATIONAL TAX PLANNING
This research examines the impact of tax treaties on international tax...
THE APPLICATION OF ANDRAGOGY IN TEACHING ENGLISH IN SENIOR SECONDARY SCHOOLS IN RIVERS STATE
BACKGROUND OF THE STUDY
The process of acquiring new knowledge, skills, and perspectives is known as le...
THE IMPACT SOCIAL INTERACTIONS ON THE PERFORMANCE OF ADULT LEARNERS
ABSTRACT
This ex-post facto research study determined the influence of social interaction on academ...
CAUSES AND EFFECT OF STUDENT’S FAILURE OF MATHEMATICS IN SENIOR SECONDARY SCHOOL CERTIFICATE EXAMINATION
Background of Study
Mathematics is an aspect of science dealing with numbers. These numbers which...
THE EFFECT OF CBN CRYPTOCURRENCY BAN ON THE ECONOMY OF NIGERIA
ABSTRACT
This study was carried out to examine the effect of CBN cryptocurrency ban on the economy of Nigeria. To achiev...
