Artificial Intelligence and Robotics Commons^™

Adapting Single-View View Synthesis With Multiplane Images For 3d Video Chat, Anurag Venkata Uppuluri

Master's Theses

Activities like one-on-one video chatting and video conferencing with multiple participants are more prevalent than ever today as we continue to tackle the pandemic. Bringing a 3D feel to video chat has always been a hot topic in Vision and Graphics communities. In this thesis, we have employed novel view synthesis in attempting to turn one-on-one video chatting into 3D. We have tuned the learning pipeline of Tucker and Snavely's single-view view synthesis paper — by retraining it on MannequinChallenge dataset — to better predict a layered representation of the scene viewed by either video chat participant at any given …

Notions Of Explainability And Evaluation Approaches For Explainable Artificial Intelligence, Giulia Vilone, Luca Longo

Articles

Explainable Artificial Intelligence (XAI) has experienced a significant growth over the last few years. This is due to the widespread application of machine learning, particularly deep learning, that has led to the development of highly accurate models that lack explainability and interpretability. A plethora of methods to tackle this problem have been proposed, developed and tested, coupled with several studies attempting to define the concept of explainability and its evaluation. This systematic review contributes to the body of knowledge by clustering all the scientific studies via a hierarchical system that classifies theories and notions related to the concept of explainability …

Plant Disease Detection Through Convolutional Neural Networks: A Survey Of Existing Literature, Best Practices, And Implementation, Kevin Label

Master's Theses

In the United States alone, common diseases spread among plants account for billions of dollars lost in crop yield each year. This issue is exacerbated in countries with less infrastructure to defend against crop epidemics, and can lead to famine and forced migration. Farmers can seek the help of plant pathology experts to defend against diseases and detect crop irregularities early on. However, access to experts can be difficult, and even those trained in the field may miss symptoms before it is too late. To assist in early disease detection, a number of papers have been released on the potential …

Determining States Of Movement In Humans Using Minimally Processed Eeg Signals And Various Classification Methods, Maurice Barnett

All Theses

Electroencephalography (EEG) is a non-invasive technique used in both clinical and research settings to record neuronal signaling in the brain. The location of an EEG signal as well as the frequencies at which its neuronal constituents fire correlate with behavioral tasks, including discrete states of motor activity. Due to the number of channels and fine temporal resolution of EEG, a dense, high-dimensional dataset is collected. Transcranial direct current stimulation (tDCS) is a treatment that has been suggested to improve motor functions of Parkinson’s disease and chronic stroke patients when stimulation occurs during a motor task. tDCS is commonly administered without …

Visualizing Features From Deep Neural Networks Trained On Alzheimer’S Disease And Few-Shot Learning Models For Alzheimer’S Disease, John Reeder

All Theses

Alzheimer’s disease is an incurable neural disease, usually affecting the elderly. The afflicted suffer from cognitive impairments that get dramatically worse at each stage. Previous research on Alzheimer’s disease analysis in terms of classification leveraged statistical models such as support vector machines. However, statistical models such as support vector machines train the from numerical data instead of medical images. Today, convolutional neural networks (CNN) are widely considered as the one which can achieve the state-of-the- art image classification performance. However, due to their black box nature, there can be reluctance amongst medical professionals for their use. On the other hand, …

Local Feature Selection For Multiple Instance Learning With Applications., Aliasghar Shahrjooihaghighi

Electronic Theses and Dissertations

Feature selection is a data processing approach that has been successfully and effectively used in developing machine learning algorithms for various applications. It has been proven to effectively reduce the dimensionality of the data and increase the accuracy and interpretability of machine learning algorithms. Conventional feature selection algorithms assume that there is an optimal global subset of features for the whole sample space. Thus, only one global subset of relevant features is learned. An alternative approach is based on the concept of Local Feature Selection (LFS), where each training sample can have its own subset of relevant features. Multiple Instance …

Rmix: Learning Risk-Sensitive Policies For Cooperative Reinforcement Learning Agents, Wei Qiu, Xinrun Wang, Runsheng Yu, Xu He, Rundong Wang, Bo An, Svetlana Obraztsova, Zinovi Rabinovich

Research Collection School Of Computing and Information Systems

Current value-based multi-agent reinforcement learning methods optimize individual Q values to guide individuals' behaviours via centralized training with decentralized execution (CTDE). However, such expected, i.e., risk-neutral, Q value is not sufficient even with CTDE due to the randomness of rewards and the uncertainty in environments, which causes the failure of these methods to train coordinating agents in complex environments. To address these issues, we propose RMIX, a novel cooperative MARL method with the Conditional Value at Risk (CVaR) measure over the learned distributions of individuals' Q values. Specifically, we first learn the return distributions of individuals to analytically calculate CVaR …

Rmm: Reinforced Memory Management For Class-Incremental Learning, Yaoyao Liu, Qianru Sun, Qianru Sun

Research Collection School Of Computing and Information Systems

Class-Incremental Learning (CIL) [38] trains classifiers under a strict memory budget: in each incremental phase, learning is done for new data, most of which is abandoned to free space for the next phase. The preserved data are exemplars used for replaying. However, existing methods use a static and ad hoc strategy for memory allocation, which is often sub-optimal. In this work, we propose a dynamic memory management strategy that is optimized for the incremental phases and different object classes. We call our method reinforced memory management (RMM), leveraging reinforcement learning. RMM training is not naturally compatible with CIL as the …

Adadeep: A Usage-Driven, Automated Deep Model Compression Framework For Enabling Ubiquitous Intelligent Mobiles, Sicong Liu, Junzhao Du, Kaiming Nan, Zimu Zhou, Hui Liu, Zhangyang Wang, Yingyan Lin

Research Collection School Of Computing and Information Systems

Recent breakthroughs in deep neural networks (DNNs) have fueled a tremendously growing demand for bringing DNN-powered intelligence into mobile platforms. While the potential of deploying DNNs on resource-constrained platforms has been demonstrated by DNN compression techniques, the current practice suffers from two limitations: 1) merely stand-alone compression schemes are investigated even though each compression technique only suit for certain types of DNN layers; and 2) mostly compression techniques are optimized for DNNs’ inference accuracy, without explicitly considering other application-driven system performance (e.g., latency and energy cost) and the varying resource availability across platforms (e.g., storage and processing capability). To this …

Ai And The Future Of Work: What We Know Today, Steven M. Miller, Thomas H. Davenport

Research Collection School Of Computing and Information Systems

To contribute to a better understanding of the contemporary realities of AI workplace deployments, the authors recently completed 29 case studies of people doing their everyday work with AI-enabled smart machines. Twenty-three of these examples were from North America, mostly in the US. Six were from Southeast Asia, mostly in Singapore. In this essay, we compare our findings on job and workplace impacts to those reported in the MIT Task Force on the Work of the Future report, as we consider that to be the most comprehensive recent study on this topic.

Solving The Vehicle Routing Problem With Simultaneous Pickup And Delivery And Occasional Drivers By Simulated Annealing, Vincent F. Yu, Grace Aloina, Panca Jodiawan, Aldy Gunawan, Tsung-Chi Huang

Research Collection School Of Computing and Information Systems

This research studies the vehicle routing problem with simultaneous pickup and delivery with an occasional driver (VRPSPDOD). VRPSPDOD is a new variant of the vehicle routing problems with simultaneous pickup and delivery (VRPSPD). Different from VRPSPD, in VRPSPDOD, occasional drivers are employed to work with regular vehicles to service customers’ pickup and delivery requests in order to minimize the total cost. We formulate a mixed integer linear programming model for VRPSPD and propose a heuristic algorithm based on simulated annealing (SA) to solve the problem. The results of comprehensive numerical experiments show that the proposed SA performs well in terms …

Building Action Sets In A Deep Reinforcement Learner, Yongzhao Wang, Arunesh Sinha, Sky C.H. Wang, Michael P. Wellman

Research Collection School Of Computing and Information Systems

In many policy-learning applications, the agent may execute a set of actions at each decision stage. Choosing among an exponential number of alternatives poses a computational challenge, and even representing actions naturally expressed as sets can be a tricky design problem. Building upon prior approaches that employ deep neural networks and iterative construction of action sets, we introduce a reward-shaping approach to apportion reward to each atomic action based on its marginal contribution within an action set, thereby providing useful feedback for learning to build these sets. We demonstrate our method in two environments where action spaces are combinatorial. Experiments …

Integration Of Blockchain Technology Into Automobiles To Prevent And Study The Causes Of Accidents, John Kim

Electronic Theses, Projects, and Dissertations

Automobile collisions occur daily. We now live in an information-driven world, one where technology is quickly evolving. Blockchain technology can change the automotive industry, the safety of the motoring public and its surrounding environment by incorporating this vast array of information. It can place safety and efficiency at the forefront to pedestrians, public establishments, and provide public agencies with pertinent information securely and efficiently. Other industries where Blockchain technology has been effective in are as follows: supply chain management, logistics, and banking. This paper reviews some statistical information regarding automobile collisions, Blockchain technology, Smart Contracts, Smart Cities; assesses the feasibility …

Integration Of Internet Of Things And Health Recommender Systems, Moonkyung Yang

Electronic Theses, Projects, and Dissertations

The Internet of Things (IoT) has become a part of our lives and has provided many enhancements to day-to-day living. In this project, IoT in healthcare is reviewed. IoT-based healthcare is utilized in remote health monitoring, observing chronic diseases, individual fitness programs, helping the elderly, and many other healthcare fields. There are three main architectures of smart IoT healthcare: Three-Layer Architecture, Service-Oriented Based Architecture (SoA), and The Middleware-Based IoT Architecture. Depending on the required services, different IoT architecture are being used. In addition, IoT healthcare services, IoT healthcare service enablers, IoT healthcare applications, and IoT healthcare services focusing on Smartwatch …

Material Handling With Embodied Loco-Manipulation, Jean Chagas Vaz

UNLV Theses, Dissertations, Professional Papers, and Capstones

Material handling is an intrinsic component of disaster response. Typically, first responders, such as firefighters and/or paramedics, must carry, push, pull, and handle objects, facilitating the transportation of goods. For many years, researchers from around the globe have sought to enable full-sized humanoid robots to perform such essential material handling tasks. This work aims to tackle current limitations of humanoids in the realm of interaction with common objects such as carts, wheelbarrows, etc. Throughout this research, many methods will be applied to ensure a stable Zero Moment Point (ZMP) trajectory to allow a robust gait while loco-manipulating a cart. The …

A Human-Embodied Drone For Dexterous Aerial Manipulation, Dongbin Kim

UNLV Theses, Dissertations, Professional Papers, and Capstones

Current drones perform a wide variety of tasks in surveillance, photography, agriculture, package delivery, etc. However, these tasks are performed passively without the use of human interaction. Aerial manipulation shifts this paradigm and implements drones with robotic arms that allow interaction with the environment rather than simply sensing it. For example, in construction, aerial manipulation in conjunction with human interaction could allow operators to perform several tasks, such as hosing decks, drill into surfaces, and sealing cracks via a drone. This integration with drones will henceforth be known as dexterous aerial manipulation.

Our recent work integrated the worker’s experience into …

Deep Learning Predicts Ebv Status In Gastric Cancer Based On Spatial Patterns Of Lymphocyte Infiltration, Baoyi Zhang, Kevin Yao, Min Xu, Jia Wu, Chao Cheng

Computer Vision Faculty Publications

EBV infection occurs in around 10% of gastric cancer cases and represents a distinct subtype, characterized by a unique mutation profile, hypermethylation, and overexpression of PD-L1. Moreover, EBV positive gastric cancer tends to have higher immune infiltration and a better prognosis. EBV infection status in gastric cancer is most commonly determined using PCR and in situ hybridization, but such a method requires good nucleic acid preservation. Detection of EBV status with histopathology images may complement PCR and in situ hybridization as a first step of EBV infection assessment. Here, we developed a deep learning-based algorithm to directly predict EBV infection …

Evaluation Of Deep Neural Network Prospr For Accurate Protein Distance Predictions On Casp14 Targets, Jacob A. Stern, Bryce Eric Hedelius, Olivia Fisher, Wendy M. Billings, Dennis Della Corte

Faculty Publications

The field of protein structure prediction has recently been revolutionized through the introduction of deep learning. The current state-of-the-art tool AlphaFold2 can predict highly accurate structures; however, it has a prohibitively long inference time for applications that require the folding of hundreds of sequences. The prediction of protein structure annotations, such as amino acid distances, can be achieved at a higher speed with existing tools, such as the ProSPr network. Here, we report on important updates to the ProSPr network, its performance in the recent Critical Assessment of Techniques for Protein Structure Prediction (CASP14) competition, and an evaluation of its …

Multi-Modal Transformers Excel At Class-Agnostic Object Detection, Muhammad Maaz, Hanoona Bangalath Rasheed, Salman Hameed Khan, Fahad Shahbaz Khan, Rao Muhammad Anwer, Ming-Hsuan Yang

Computer Vision Faculty Publications

What constitutes an object? This has been a longstanding question in computer vision. Towards this goal, numerous learning-free and learning-based approaches have been developed to score objectness. However, they generally do not scale well across new domains and for unseen objects. In this paper, we advocate that existing methods lack a top-down supervision signal governed by human-understandable semantics. To bridge this gap, we explore recent Multi-modal Vision Transformers (MViT) that have been trained with aligned image-text pairs. Our extensive experiments across various domains and novel objects show the state-of-the-art performance of MViTs to localize generic objects in images. Based on …

Situate: An Agent-Based System For Situation Recognition, Max Henry Quinn

Dissertations and Theses

Computer vision and machine learning systems have improved significantly in recent years, largely based on the development of deep learning systems, leading to impressive performance on object detection tasks. Understanding the content of images is considerably more difficult. Even simple situations, such as "a handshake", "walking the dog", "a game of ping-pong", or "people waiting for a bus", present significant challenges. Each consists of common objects, but are not reliably detectable as a single entity nor through the simple co-occurrence of their parts.

In this dissertation, toward the goal of developing machine learning systems that demonstrate properties associated with understanding, …