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 5000
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 appraisal of the challenges in assessing students’ moral development in secondary schools in Yola South LGA, Adamawa State
Background of the Study
The assessment of students’ moral development in secondary schools has become a subject of critical inquiry...
THE USE OF ACCOUNTING TECHNIQUES IN NIGERIAN SMES
ABSTRACT
There have been the recent calls for additional research in order to enhance the understanding of the ado...
The effect of forced marriages on women's rights in Argungu, Kebbi State.
Background of the Study:
Forced marriage is a widespread violation of human rights, particularly affecting young girls and...
The Effect of Sexual Education Programs on Adolescent HIV Prevention in Cross River State
Background of the Study
Sexual education programs have been widely recognized as essential components of H...
Effectiveness of Local Radio in Promoting Community-Led Development Initiatives: A Study of Kaura Local Government Area, Kaduna State
Chapter One: Introduction
1.1 Background of the Study
Community-led development initiatives are essential for sustainable devel...
THE IMPORTANCE OF STUDYING CO-OPERATIVE ECONOMICS AND MANAGEMENT
Introduction
Co-operative Economics and Management in tertiary institutions help to improve the co-oper...
An Appraisal of Constitutional Law and the Protection of Media Practitioners: A Study of Gombe Local Government Area, Gombe State
Background of the Study
In Nigeria, media freedom is enshrined in the Constitu...
DESIGN AND CONTRUCTION OF AUTOMATIC PHASE SELECTOR
BACKGROUND
The intelligent phase selector is a system that is capable of comparing three phases and switching automatically to any of the...
An Examination of Code Switching Practices in Nigerian Digital Communication: A Case Study of WhatsApp Group Chats
Background of the study
Code‑switching—the practice of alternating between languages within a conversation—is a...
An Assessment of Brand Equity and Consumer Trust: A Study of Dangote Cement in Kano State
Background of the Study
Brand equity and consumer trust are critical components of a company's success, especially i...
