Artificial Intelligence and Robotics Commons^™

Deep Learning For Anomaly Detection, Guansong Pang, Charu Aggarwal, Chunhua Shen, Nicu Sebe

Research Collection School Of Computing and Information Systems

A nomaly detection aims at identifying data points which are rare or significantly different from the majority of data points. Many techniques are explored to build highly efficient and effective anomaly detection systems, but they are confronted with many difficulties when dealing with complex data, such as failing to capture intricate feature interactions or extract good feature representations. Deep-learning techniques have shown very promising performance in tackling different types of complex data in a broad range of tasks/problems, including anomaly detection. To address this new trend, we organized this Special Issue on Deep Learning for Anomaly Detection to cover the …

Challenges In Migrating Imperative Deep Learning Programs To Graph Execution: An Empirical Study, Tatiana Castro Vélez, Raffi Khatchadourian, Mehdi Bagherzadeh, Anita Raja

Publications and Research

Efficiency is essential to support responsiveness w.r.t. ever-growing datasets, especially for Deep Learning (DL) systems. DL frameworks have traditionally embraced deferred execution-style DL code that supports symbolic, graph-based Deep Neural Network (DNN) computation. While scalable, such development tends to produce DL code that is error-prone, non-intuitive, and difficult to debug. Consequently, more natural, less error-prone imperative DL frameworks encouraging eager execution have emerged at the expense of run-time performance. While hybrid approaches aim for the "best of both worlds," the challenges in applying them in the real world are largely unknown. We conduct a data-driven analysis of challenges—and resultant bugs—involved …

Challenges In Migrating Imperative Deep Learning Programs To Graph Execution: An Empirical Study, Tatiana Castro Vélez, Raffi Khatchadourian, Mehdi Bagherzadeh, Anita Raja

Publications and Research

Efficiency is essential to support responsiveness w.r.t. ever-growing datasets, especially for Deep Learning (DL) systems. DL frameworks have traditionally embraced deferred execution-style DL code that supports symbolic, graph-based Deep Neural Network (DNN) computation. While scalable, such development tends to produce DL code that is error-prone, non-intuitive, and difficult to debug. Consequently, more natural, less error-prone imperative DL frameworks encouraging eager execution have emerged at the expense of run-time performance. While hybrid approaches aim for the "best of both worlds," the challenges in applying them in the real world are largely unknown. We conduct a data-driven analysis of challenges—and resultant bugs—involved …

Identifying Noisy Labels In The Ground Truth Of Eating Episodes Self-Reported By Button Press On A Wrist-Worn Device, Tianyi Zhang

All Theses

This thesis considers the problem of identifying noisy labels in the ground truth of eating episodes (meals, snacks) as self-reported by participants collecting data in the wild. Participants wore a smartwatch-like device that tracked their wrist motion all day. They were instructed to press a button on the device at the start and end of each eating episode. The device and instructions were designed to be as simple to use as possible, but post-review of the ground truth provided by participants revealed a strong likelihood that a significant portion of the button presses may contain errors. For example, an error …

Cell Tracking At Low Frame Rate Using Deep Learning And Bayesian Integration, Xiang Zhang

All Dissertations

Tracking cells over time is a fundamental task in live-cell imaging, and often requires costly manual analysis if images are not acquired with high enough frame rate. Acquiring high frame rate images, however, can limit the number of conditions explored and cells analyzed, and contribute to photobleaching, which makes fluorophores dimmer and phototoxicity, which affects cell health and renders the resulting data unusable.

Assuming a relatively high frame rate in image acquisition, state-of-the-art cell tracking approaches rely on either spatial proximity or morphological similarity to link cells in consecutive frames. The problem is that, at low frame rate, both approaches …

Indoor Localization Using Solar Cells, Hamada Rizk, Dong Ma, Mahbub Hassan, Moustafa Youssef

Research Collection School Of Computing and Information Systems

The development of the Internet of Things (IoT) opens the doors for innovative solutions in indoor positioning systems. Recently, light-based positioning has attracted much attention due to the dense and pervasive nature of light sources (e.g., Light-emitting Diode lighting) in indoor environments. Nevertheless, most existing solutions necessitate carrying a high-end phone at hand in a specific orientation to detect the light intensity with the phone's light sensing capability (i.e., light sensor or camera). This limits the ease of deployment of these solutions and leads to drainage of the phone battery. We propose PVDeepLoc, a device-free light-based indoor localization system that …

Deep Learning Object-Based Detection Of Manufacturing Defects In X-Ray Inspection Imaging, Juan C. Parducci

Mechanical & Aerospace Engineering Theses & Dissertations

Current analysis of manufacturing defects in the production of rims and tires via x-ray inspection at an industry partner’s manufacturing plant requires that a quality control specialist visually inspect radiographic images for defects of varying sizes. For each sample, twelve radiographs are taken within 35 seconds. Some defects are very small in size and difficult to see (e.g., pinholes) whereas others are large and easily identifiable. Implementing this quality control practice across all products in its human-effort driven state is not feasible given the time constraint present for analysis.

This study aims to identify and develop an object detector capable …

Radar Remote Sensing Data Augmentation Method Based On Generative Adversarial Network, Xu Kang, Xiaofeng Zhang

Journal of System Simulation

Abstract: In the research field of radar remote sensing, both the completeness and diversity of radar data samples cannot meet the requirement of effective training of deep learning models, and the models are prone to over-fitting, which significantly limits the wide application of deep learning techniques in this field. Targeting on the needs of intelligent application in radar remote sensing, a microwave imaging radar suited data augmentation method is proposed to solve the issue of insufficient radar data samples by leveraging the general framework of generative adversarial network. Aiming at the features of radar samples being not obvious, the label …

Visual Attention Methods In Deep Learning: An In-Depth Survey, Mohammed Hassanin, Anwar Saeed, Ibrahim Radwan, Fahad Shahbaz Khan, Ajmal Mian

Computer Vision Faculty Publications

Inspired by the human cognitive system, attention is a mechanism that imitates the human cognitive awareness about specific information, amplifying critical details to focus more on the essential aspects of data. Deep learning has employed attention to boost performance for many applications. Interestingly, the same attention design can suit processing different data modalities and can easily be incorporated into large networks. Furthermore, multiple complementary attention mechanisms can be incorporated in one network. Hence, attention techniques have become extremely attractive. However, the literature lacks a comprehensive survey specific to attention techniques to guide researchers in employing attention in their deep models. …

Intra-Hour Solar Forecasting Using Cloud Dynamics Features Extracted From Ground-Based Infrared Sky Images, Guillermo Terrén-Serrano

Electrical and Computer Engineering ETDs

Due to the increasing use of photovoltaic systems, power grids are vulnerable to the projection of shadows from moving clouds. An intra-hour solar forecast provides power grids with the capability of automatically controlling the dispatch of energy, reducing the additional cost for a guaranteed, reliable supply of energy (i.e., energy storage). This dissertation introduces a novel sky imager consisting of a long-wave radiometric infrared camera and a visible light camera with a fisheye lens. The imager is mounted on a solar tracker to maintain the Sun in the center of the images throughout the day, reducing the scattering effect produced …

Toward Suicidal Ideation Detection With Lexical Network Features And Machine Learning, Ulya Bayram, William Lee, Daniel Santel, Ali Minai, Peggy Clark, Tracy Glauser, John Pestian

Northeast Journal of Complex Systems (NEJCS)

In this study, we introduce a new network feature for detecting suicidal ideation from clinical texts and conduct various additional experiments to enrich the state of knowledge. We evaluate statistical features with and without stopwords, use lexical networks for feature extraction and classification, and compare the results with standard machine learning methods using a logistic classifier, a neural network, and a deep learning method. We utilize three text collections. The first two contain transcriptions of interviews conducted by experts with suicidal (n=161 patients that experienced severe ideation) and control subjects (n=153). The third collection consists of interviews conducted by experts …

Comai: Enabling Lightweight, Collaborative Intelligence By Retrofitting Vision Dnns, Kasthuri Jayarajah, Dhanuja Wanniarachchige, Tarek Abdelzaher, Archan Misra

Research Collection School Of Computing and Information Systems

While Deep Neural Network (DNN) models have transformed machine vision capabilities, their extremely high computational complexity and model sizes present a formidable deployment roadblock for AIoT applications. We show that the complexity-vs-accuracy-vs-communication tradeoffs for such DNN models can be significantly addressed via a novel, lightweight form of “collaborative machine intelligence” that requires only runtime changes to the inference process. In our proposed approach, called ComAI, the DNN pipelines of different vision sensors share intermediate processing state with one another, effectively providing hints about objects located within their mutually-overlapping Field-of-Views (FoVs). CoMAI uses two novel techniques: (a) a secondary shallow ML …

Energy-Based Latent Aligner For Incremental Learning, K.J. Joseph, Salman Khan, Fahad Shahbaz Khan, Rao Muhammad Anwer, Vineeth N. Balasubramanian

Computer Vision Faculty Publications

Deep learning models tend to forget their earlier knowledge while incrementally learning new tasks. This behavior emerges because the parameter updates optimized for the new tasks may not align well with the updates suitable for older tasks. The resulting latent representation mismatch causes forgetting. In this work, we propose ELI: Energy-based Latent Aligner for Incremental Learning, which first learns an energy manifold for the latent representations such that previous task latents will have low energy and the current task latents have high energy values. This learned manifold is used to counter the representational shift that happens during incremental learning. The …

Deep Learning Techniques For Diabetic Retinopathy Classification: A Survey, Mohammad Z. Atwany, Abdulwahab H. Sahyoun, Mohammad Yaqub

Computer Vision Faculty Publications

Diabetic Retinopathy (DR) is a degenerative disease that impacts the eyes and is a consequence of Diabetes mellitus, where high blood glucose levels induce lesions on the eye retina. Diabetic Retinopathy is regarded as the leading cause of blindness for diabetic patients, especially the working-age population in developing nations. Treatment involves sustaining the patient's current grade of vision since the disease is irreversible. Early detection of Diabetic Retinopathy is crucial in order to sustain the patient's vision effectively. The main issue involved with DR detection is that the manual diagnosis process is very time, money, and effort consuming and involves …

Deep Learning For Anomaly Detection: A Review, Guansong Pang, Chunhua Shen, Longbing Cao, Anton Van Den Hengel

Research Collection School Of Computing and Information Systems

Anomaly detection, a.k.a. outlier detection or novelty detection, has been a lasting yet active research area in various research communities for several decades. There are still some unique problem complexities and challenges that require advanced approaches. In recent years, deep learning enabled anomaly detection, i.e., deep anomaly detection, has emerged as a critical direction. This article surveys the research of deep anomaly detection with a comprehensive taxonomy, covering advancements in 3 high-level categories and 11 fine-grained categories of the methods. We review their key intuitions, objective functions, underlying assumptions, advantages, and disadvantages and discuss how they address the aforementioned challenges. …

Heterogeneous Attentions For Solving Pickup And Delivery Problem Via Deep Reinforcement Learning, Jingwen Li, Liang Xin, Zhiguang Cao, Andrew Lim, Wen Song, Jie Zhang

Research Collection School Of Computing and Information Systems

Recently, there is an emerging trend to apply deep reinforcement learning to solve the vehicle routing problem (VRP), where a learnt policy governs the selection of next node for visiting. However, existing methods could not handle well the pairing and precedence relationships in the pickup and delivery problem (PDP), which is a representative variant of VRP. To address this challenging issue, we leverage a novel neural network integrated with a heterogeneous attention mechanism to empower the policy in deep reinforcement learning to automatically select the nodes. In particular, the heterogeneous attention mechanism specifically prescribes attentions for each role of the …

An Ensemble Approach For Patient Prognosis Of Head And Neck Tumor Using Multimodal Data, Numan Saeed, Roba Al Majzoub, Ikboljon Sobirov, Mohammad Yaqub

Computer Vision Faculty Publications

Accurate prognosis of a tumor can help doctors provide a proper course of treatment and, therefore, save the lives of many. Tradi-tional machine learning algorithms have been eminently useful in crafting prognostic models in the last few decades. Recently, deep learning algorithms have shown significant improvement when developing diag-nosis and prognosis solutions to different healthcare problems. However, most of these solutions rely solely on either imaging or clinical data. Utilizing patient tabular data such as demographics and patient med-ical history alongside imaging data in a multimodal approach to solve a prognosis task has started to gain more interest recently and …

Deep-Precognitive Diagnosis: Preventing Future Pandemics By Novel Disease Detection With Biologically-Inspired Conv-Fuzzy Network, Aviral Chharia, Rahul Upadhyay, Vinay Kumar, Chao Cheng, Jing Zhang, Tianyang Wang, Min Xu

Computer Vision Faculty Publications

Deep learning-based Computer-Aided Diagnosis has gained immense attention in recent years due to its capability to enhance diagnostic performance and elucidate complex clinical tasks. However, conventional supervised deep learning models are incapable of recognizing novel diseases that do not exist in the training dataset. Automated early-stage detection of novel infectious diseases can be vital in controlling their rapid spread. Moreover, the development of a conventional CAD model is only possible after disease outbreaks and datasets become available for training (viz. COVID-19 outbreak). Since novel diseases are unknown and cannot be included in training data, it is challenging to recognize them …

Survey On Self-Supervised Representation Learning Using Image Transformations, Muhammad Ali, Sayed Hashim

Student Publications

Deep neural networks need huge amount of training data, while in real world there is a scarcity of data available for training purposes. To resolve these issues, self-supervised learning (SSL) methods are used. SSL using geometric transformations (GT) is a simple yet powerful technique used in unsupervised representation learning. Although multiple survey papers have reviewed SSL techniques, there is none that only focuses on those that use geometric transformations. Furthermore, such methods have not been covered in depth in papers where they are reviewed. Our motivation to present this work is that geometric transformations have shown to be powerful supervisory …

Cocoa: Context-Conditional Adaptation For Recognizing Unseen Classes In Unseen Domains, Puneet Mangla, Shivam Chandhok, Vineeth N. Balasubramanian, Fahad Shahbaz Khan

Computer Vision Faculty Publications

Recent progress towards designing models that can generalize to unseen domains (i.e domain generalization) or unseen classes (i.e zero-shot learning) has embarked interest towards building models that can tackle both domain-shift and semantic shift simultaneously (i.e zero-shot domain generalization). For models to generalize to unseen classes in unseen domains, it is crucial to learn feature representation that preserves class-level (domain-invariant) as well as domain-specific information. Motivated from the success of generative zero-shot approaches, we propose a feature generative framework integrated with a COntext COnditional Adaptive (COCOA) Batch-Normalization layer to seamlessly integrate class-level semantic and domain-specific information. The generated visual features …

Diagnosis Of Polypoidal Choroidal Vasculopathy From Fluorescein Angiography Using Deep Learning, Yu-Yeh Tsai, Wei-Yang Ling, Shih-Jen Chen, Paisan Ruamviboonsuk, Cheng-Ho King, Chia-Ling Tsai

Publications and Research

Purpose: To differentiate polypoidal choroidal vasculopathy (PCV) from choroidal neovascularization (CNV) and to determine the extent of PCV from fluorescein angiography (FA) using attention-based deep learning networks.

Methods: We build two deep learning networks for diagnosis of PCV using FA, one for detection and one for segmentation. Attention-gated convolutional neural network (AG-CNN) differentiates PCV from other types of wet age-related macular degeneration. Gradient-weighted class activation map (Grad-CAM) is generated to highlight important regions in the image for making the prediction, which offers explainability of the network. Attention-gated recurrent neural network (AG-PCVNet) for spatiotemporal prediction is applied for segmentation …

Diagnosis Of Polypoidal Choroidal Vasculopathy From Fluorescein Angiography Using Deep Learning, Yu-Yeh Tsai, Wei-Yang Lin, Shih-Jen Chen, Paisan Ruamviboonsuk, Cheng-Ho King, Chia-Ling Tsai

Publications and Research

Purpose: To differentiate polypoidal choroidal vasculopathy (PCV) from choroidal neovascularization (CNV) and to determine the extent of PCV from fluorescein angiography (FA) using attention-based deep learning networks.

Methods: We build two deep learning networks for diagnosis of PCV using FA, one for detection and one for segmentation. Attention-gated convolutional neural network (AG-CNN) differentiates PCV from other types of wet age-related macular degeneration. Gradient-weighted class activation map (Grad-CAM) is generated to highlight important regions in the image for making the prediction, which offers explainability of the network. Attention-gated recurrent neural network (AG-PCVNet) for spatiotemporal prediction is applied for segmentation of PCV. …

Understanding Deep Learning - Challenges And Prospects, Niha Adnan, Fahad Umer

Department of Surgery Docs

The developments in Artificial Intelligence have been on the rise since its advent. The advancements in this field have been the innovative research area across a wide range of industries, making its incorporation in dentistry inevitable. Artificial Intelligence techniques are making serious progress in the diagnostic and treatment planning aspects of dental clinical practice. This will ultimately help in the elimination of subjectivity and human error that are often part of radiographic interpretations, and will improve the overall efficiency of the process. The various types of Artificial Intelligence algorithms that exist today make the understanding of their application quite complex. …

Deep Graph-Level Anomaly Detection By Glocal Knowledge Distillation, Rongrong Ma, Guansong Pang, Ling Chen, Anton Van Den Hengel

Research Collection School Of Computing and Information Systems

Graph-level anomaly detection (GAD) describes the problem of detecting graphs that are abnormal in their structure and/or the features of their nodes, as compared to other graphs. One of the challenges in GAD is to devise graph representations that enable the detection of both locally- and globally-anomalous graphs, i.e., graphs that are abnormal in their fine-grained (node-level) or holistic (graph-level) properties, respectively. To tackle this challenge we introduce a novel deep anomaly detection approach for GAD that learns rich global and local normal pattern information by joint random distillation of graph and node representations. The random distillation is achieved by …

Automatic Segmentation Of Head And Neck Tumor: How Powerful Transformers Are?, Ikboljon Sobirov, Otabek Nazarov, Hussain Alasmawi, Mohammad Yaqub

Computer Vision Faculty Publications

Cancer is one of the leading causes of death worldwide, and head and neck (H&N) cancer is amongst the most prevalent types. Positron emission tomography and computed tomography are used to detect and segment the tumor region. Clinically, tumor segmentation is extensively time-consuming and prone to error. Machine learning, and deep learning in particular, can assist to automate this process, yielding results as accurate as the results of a clinician. In this research study, we develop a vision transformers-based method to automatically delineate H&N tumor, and compare its results to leading convolutional neural network (CNN)-based models. We use multi-modal data …

Is It Possible To Predict Mgmt Promoter Methylation From Brain Tumor Mri Scans Using Deep Learning Models?, Numan Saeed, Shahad Hardan, Kudaibergen Abutalip, Mohammad Yaqub

Computer Vision Faculty Publications

Glioblastoma is a common brain malignancy that tends to occur in older adults and is almost always lethal. The effectiveness of chemotherapy, being the standard treatment for most cancer types, can be improved if a particular genetic sequence in the tumor known as MGMT promoter is methylated. However, to identify the state of the MGMT promoter, the conventional approach is to perform a biopsy for genetic analysis, which is time and effort consuming. A couple of recent publications proposed a connection between the MGMT promoter state and the MRI scans of the tumor and hence suggested the use of deep …

Challenges In Covid-19 Chest X-Ray Classification: Problematic Data Or Ineffective Approaches?, Muhammad Ridzuan, Ameera Ali Bawazir, Ivo Gollini Navarrete, Ibrahim Almakky, Mohammad Yaqub

Computer Vision Faculty Publications

The value of quick, accurate, and confident diagnoses cannot be undermined to mitigate the effects of COVID-19 infection, particularly for severe cases. Enormous effort has been put towards developing deep learning methods to classify and detect COVID-19 infections from chest radiography images. However, recently some questions have been raised surrounding the clinical viability and effectiveness of such methods. In this work, we carry out extensive experiments on a large COVID-19 chest X-ray dataset to investigate the challenges faced with creating reliable solutions from both the data and machine learning perspectives. Accordingly, we offer an in-depth discussion into the challenges faced …

Study On Prediction Of Crystal Properties Based On Deep Learning, Buwei Wang, Wang Min, Fan Qian, Ya'nan Wang, Hanwen Zhang, Yunliang Yue

Journal of System Simulation

Abstract: Predicting crystal properties using traditional machine learning methods requires complex feature engineering. In order to bypass time-consuming feature engineering, element network (ElemNet), representation learning from stoichiometry (Roost), compositionally-restricted attention-based network (CrabNet) and crystal graph convolution neural network (CGCNN) based on deep learning technology are used to simulate the formation energy, total energy per atom, band gap, and Fermi energy of crystal. The residual learning is introduced into CGCNN, and a crystal graph convolution residual neural network (CGCRN) is proposed. In the CGCRN, the number of hidden layers and the number of nodes in the hidden layers are increased, …

Brief Review On Applying Reinforcement Learning To Job Shop Scheduling Problems, Xiaohan Wang, Zhang Lin, Ren Lei, Kunyu Xie, Kunyu Wang, Ye Fei, Chen Zhen

Journal of System Simulation

Abstract: Reinforcement Learning (RL) achieves lower time response and better model generalization in Job Shop Scheduling Problem (JSSP). To explain the current overall research status of JSSP based on RL, summarize the current scheduling framework based on RL, and lay the foundation for follow-up research, the backgrounds of JSSP and RL are introduced. Two simulation techniques commonly used in JSSP are analyzed and two commonly used frameworks for RL to solve JSSP are given. In addition, some existing challenges are pointed out, and related research progress is introduced from three aspects: direct scheduling, feature representation-based scheduling, and parameter search-based scheduling.

Variety Recognition Based On Deep Learning And Double-Sided Characteristics Of Maize Kernel, Feng Xiao, Zhang Hui, Zhou Rui, Qiao Lu, Wei Dong, Dandan Li, Yuyao Zhang, Guoqing Zheng

Journal of System Simulation

Abstract: In order to construct a maize kernel variety recognition model with high recognition accuracy and suitable for mobile phone application, a mobile phone is used to obtain maize kernel double-sided (embryonic and non-embryonic) images. Based on the lightweight convolutional neural network MobileNetV2 and transfer learning, a maize kernel image variety recognition model is constructed. In view of the existing research methods are mainly for single-sided recognition of maize kernel variety, the performance of single-sided and double-sided characteristics modeling and recognition is compared. The results show that the double-sided recognition accuracy of maize kernel double-sided characteristics modeling is 99.83%, which …