Digital Commons Network^™

An Application Of Deep-Learning To Understand Human Perception Of Art, Sanjana Kapisthalam

Theses

Eye movement patterns are known to differ when looking at stimuli given a different task, but less is known about how these patterns change as a function of expertise. When a particular visual pattern is viewed, a particular sequence of eye movements are executed and this sequence is defined as scanpath. In this work we made an attempt to answer the question, “Do art novices and experts look at paintings differently?” If they do, we should be able to discriminate between the two groups using machine learning applied to their scanpaths. This can be done using algorithms for Multi-Fixation Pattern …

Increasing Revenue By Applying Machine Learning To Congestion Management In Sdn, Nabarun Jana

Theses

With the advent of 5G, IoT and 4k videos, online gaming, movie streaming and other data intensive applications, the demand for data is sky rocketing. Due to this surge in data, the load on the network increases. This heightened network load causes degradation in network performance. Which can lead to the customer Service Provider (CSP)s loosing revenue if the Service Level Agreement (SLA) are not met.

This report describes how machine learning techniques such as tit for tat can be applied to telecom networks. Machine learning applied to telecom networks help detect congestion and maintain SLAs while increasing yield (revenue). …

Predictive Models In Brain Connectivity Analysis, Guenadie Nibbs

Theses

Neuroscience have been the field with most significant contributions to the study of the human brain. The development of new techniques for image acquisition has made possible the improvement of extracting quality information of brain activity. Utilizing functional MRIs, is possible to measure brain activity based on changes of the oxygen level in the blood at certain period of time. This imaging data is transformed into numerical values using a software that maps all the information into a data object. Taking advantage of the availability of functional connectivity information of the human brain, the present study shows a widespread process …

Multi-Modal Deep Learning To Understand Vision And Language, Shagan Sah

Theses

Developing intelligent agents that can perceive and understand the rich visual world around us has been a long-standing goal in the field of artificial intelligence. In the last few years, significant progress has been made towards this goal and deep learning has been attributed to recent incredible advances in general visual and language understanding. Convolutional neural networks have been used to learn image representations while recurrent neural networks have demonstrated the ability to generate text from visual stimuli. In this thesis, we develop methods and techniques using hybrid convolutional and recurrent neural network architectures that connect visual data and natural …

Embedded Cyclegan For Shape-Agnostic Image-To-Image Translation, Ram Longman

Theses

Image-to-Image translation is the task of translating images between domains while maintaining the identity of the images. The task can be used for entertainment purposes and applications, data augmentation, semantic image segmentation, and more. Generative Adversarial Networks (GANs), and in particular Conditional GANs have recently shown incredible success in image-to-image translation and semantic manipulation. However, such methods require paired data, meaning that an image must have ground-truth translations across domains. Cycle-consistent GANs solve this problem by using unpaired data. Such methods work well for translations that involve color and texture changes but fail when shape changes are required. This research …

Low-Shot Learning For The Semantic Segmentation Of Remote Sensing Imagery, Ronald Kemker

Theses

Deep-learning frameworks have made remarkable progress thanks to the creation of large annotated datasets such as ImageNet, which has over one million training images. Although this works well for color (RGB) imagery, labeled datasets for other sensor modalities (e.g., multispectral and hyperspectral) are minuscule in comparison. This is because annotated datasets are expensive and man-power intensive to complete; and since this would be impractical to accomplish for each type of sensor, current state-of-the-art approaches in computer vision are not ideal for remote sensing problems. The shortage of annotated remote sensing imagery beyond the visual spectrum has forced researchers to embrace …

Novel Detection And Analysis Using Deep Variational Autoencoders, Tucker B. Graydon

Theses

This paper presents a Novel Identification System which uses generative modeling techniques and Gaussian Mixture Models (GMMs) to identify the main process variables involved in a novel event from multivariate data. Features are generated and subsequently dimensionally reduced by using a Variational Autoencoder (VAE) supplemented by a denoising criterion and a β disentangling method. The GMM parameters are learned using the Expectation Maximization(EM) algorithm on features collected from only normal operating conditions. A one-class classification is achieved by thresholding the likelihoods by a statistically derived value. The Novel Identification method is verified as a detection method on existing Radio Frequency …

Contractive Autoencoding For Hierarchical Temporal Memory And Sparse Distributed Representation Binding, Luke G. Boudreau

Theses

Hierarchical Temporal Memory is a brain inspired memory prediction framework modeled after the uniform structure and connectivity of pyramidal neurons found in the human neocortex. Similar to the neocortex, Hierarchical Temporal Memory processes spatiotemporal information for anomaly detection and prediction. A critical component in the Hierarchical Temporal Memory algorithm is the Spatial Pooler, which is responsible for processing feedforward data into sparse distributed representations.

This study addresses three fundamental research questions for Hierarchical Temporal Memory algorithms. What are the metrics for understanding the semantic content of sparse distributed representations? The semantic content and relationships between representations was visualized with uniqueness …

A Satellite-Based Climatology Of Transverse Cirrus Band Occurrence Using Deep Learning, Jeffrey Miller

Theses

Transverse cirrus bands (TCBs) occur in conjunction with meteorological features such as jet streaks and hurricanes and are often used as a proxy for clear-air turbulence. This study examines the viability of using a convolutional neural network to detect TCBs in satellite imagery. The Visual Geometry Group-16 (VGG-16) network architecture developed for general-purpose image classication was adapted for TCB detection using the transfer-learning approach. The modied network successfully detected (94% accuracy) the presence of TCBs in NASA MODIS and VIIRS true-color satellite browse imagery and outperformed a random forest classier (84% accuracy) trained on the same dataset. The CNN was …