Artificial Intelligence and Robotics Commons^™

Using A Bert-Based Ensemble Network For Abusive Language Detection, Noah Ballinger

Computer Science and Computer Engineering Undergraduate Honors Theses

Over the past two decades, online discussion has skyrocketed in scope and scale. However, so has the amount of toxicity and offensive posts on social media and other discussion sites. Despite this rise in prevalence, the ability to automatically moderate online discussion platforms has seen minimal development. Recently, though, as the capabilities of artificial intelligence (AI) continue to improve, the potential of AI-based detection of harmful internet content has become a real possibility. In the past couple years, there has been a surge in performance on tasks in the field of natural language processing, mainly due to the development of …

Designing A Digital Interactive Emotion Measure (Diem) For Digital Media: Theoretical Foundations And Validation Protocols, Celeste Sangiorgio, Cassandra Berbary, Cory Crane, Caroline Easton

Frameless

Awareness of emotions is often a treatment target in psychotherapy, but it is difficult to assess emotions due to ambiguity in measurement or scale design. Lack of clarity in scale design may increase risk that participant interpretations of scale items may not align with emotion constructs those scales were designed to capture. Furthermore, emphasis on verbal or written cues leads to low scientific representation of patients who cannot read emotion scales (e.g., low literacy). Touch-screen applications provide a unique opportunity to create a visual emotion measure which has low barriers but can be used to assess a high level of …

Novel 360-Degree Camera, Ian Gauger, Andrew Kurtz, Zakariya Niazi

Frameless

Circle Optics is developing novel technology for low-parallax, real time, panoramic image capture using an integrated array of multiple adjacent polygonal-edged cameras. This technology can be optimized and deployed for a variety of markets, including cinematic VR. Circle Optics’ existing prototype, Hydra Alpha, will be demonstrated.

Warehouse And Logistics: Smart Picking With Vuzix Smart Glasses, Elise Hemink

Frameless

Vuzix is an industry leader in augmented reality (AR) technology. We provide innovative products to an array of industries, a few being defense, security, enterprise, and consumers. Our AR technology provides a perfect balance of engagement in the digital and real worlds thanks to their innovative optics, AI apps and 5G capability.

Creating A Virtual Reality Experience In Service To A Non-Profit Agency, Frank Deese, Susan Lakin, Isabelle Anderson

Frameless

In the summer of 2018, RIT Professors Susan Lakin and Frank Deese discussed with the principal officers of the Society for the Protection and Care of Children (SPCC) in Rochester how the new technology of Virtual Reality might be used to not only impart information to viewers, but generate empathy for those receiving services from the organization as well as those performing those services. Their ultimate goal was to create an experience that could be viewed with VR headsets at fundraising events and on a website using low-cost Google Cardboard.

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 …

Debiasing Nlu Models Via Causal Intervention And Counterfactual Reasoning, Bing Tian, Yixin Cao, Yong Zhang, Chunxiao Xing

Research Collection School Of Computing and Information Systems

Recent studies have shown that strong Natural Language Understanding (NLU) models are prone to relying on annotation biases of the datasets as a shortcut, which goes against the underlying mechanisms of the task of interest. To reduce such biases, several recent works introduce debiasing methods to regularize the training process of targeted NLU models. In this paper, we provide a new perspective with causal inference to fnd out the bias. On the one hand, we show that there is an unobserved confounder for the natural language utterances and their respective classes, leading to spurious correlations from training data. To remove …

Deconfounded Visual Grounding, Jianqiang Huang, Yu Qin, Jiaxin Qi, Qianru Sun, Hanwang Zhang

Research Collection School Of Computing and Information Systems

We focus on the confounding bias between language and location in the visual grounding pipeline, where we find that the bias is the major visual reasoning bottleneck. For example, the grounding process is usually a trivial languagelocation association without visual reasoning, e.g., grounding any language query containing sheep to the nearly central regions, due to that most queries about sheep have groundtruth locations at the image center. First, we frame the visual grounding pipeline into a causal graph, which shows the causalities among image, query, target location and underlying confounder. Through the causal graph, we know how to break the …

State Graph Reasoning For Multimodal Conversational Recommendation, Yuxia Wu, Lizi Liao, Gangyi Zhang, Wenqiang Lei, Guoshuai Zhao, Xueming Qian, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Conversational recommendation system (CRS) attracts increasing attention in various application domains such as retail and travel. It offers an effective way to capture users’ dynamic preferences with multi-turn conversations. However, most current studies center on the recommendation aspect while over-simplifying the conversation process. The negligence of complexity in data structure and conversation flow hinders their practicality and utility. In reality, there exist various relationships among slots and values, while users’ requirements may dynamically adjust or change. Moreover, the conversation often involves visual modality to facilitate the conversation. These actually call for a more advanced internal state representation of the dialogue …

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 …

Multi-User Eye-Tracking, Bhanuka Mahanama

Computer Science Faculty Publications

The human gaze characteristics provide informative cues on human behavior during various activities. Using traditional eye trackers, assessing gaze characteristics in the wild requires a dedicated device per participant and therefore is not feasible for large-scale experiments. In this study, we propose a commodity hardware-based multi-user eye-tracking system. We leverage the recent advancements in Deep Neural Networks and large-scale datasets for implementing our system. Our preliminary studies provide promising results for multi-user eye-tracking on commodity hardware, providing a cost-effective solution for large-scale studies.

Cross-Modal Food Retrieval: Learning A Joint Embedding Of Food Images And Recipes With Semantic Consistency And Attention Mechanism, Hao Wang, Doyen Sahoo, Chenghao Liu, Ke Shu, Palakorn Achananuparp, Ee-Peng Lim, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

Food retrieval is an important task to perform analysis of food-related information, where we are interested in retrieving relevant information about the queried food item such as ingredients, cooking instructions, etc. In this paper, we investigate cross-modal retrieval between food images and cooking recipes. The goal is to learn an embedding of images and recipes in a common feature space, such that the corresponding image-recipe embeddings lie close to one another. Two major challenges in addressing this problem are 1) large intra-variance and small inter-variance across cross-modal food data; and 2) difficulties in obtaining discriminative recipe representations. To address these …

Comparative Analysis Of Rgb-Based Eye-Tracking For Large-Scale Human-Machine Applications, Brett Thaman, Trung Cao

Posters-at-the-Capitol

Gaze tracking has become an established technology that enables using an individual’s gaze as an input signal to support a variety of applications in the context of Human-Computer Interaction. Gaze tracking primarily relies on sensing devices such as infrared (IR) cameras. Nevertheless, in the recent years, several attempts have been realized at detecting gaze by acquiring and processing images acquired from standard RGB cameras. Nowadays, there are only a few publicly available open-source libraries and they have not been tested extensively. In this paper, we present the result of a comparative analysis that studied a commercial eye-tracking device using IR …

Contrastive Learning For Unsupervised Auditory Texture Models, Christina Trexler

Computer Science and Computer Engineering Undergraduate Honors Theses

Sounds with a high level of stationarity, also known as sound textures, have perceptually relevant features which can be captured by stimulus-computable models. This makes texture-like sounds, such as those made by rain, wind, and fire, an appealing test case for understanding the underlying mechanisms of auditory recognition. Previous auditory texture models typically measured statistics from auditory filter bank representations, and the statistics they used were somewhat ad-hoc, hand-engineered through a process of trial and error. Here, we investigate whether a better auditory texture representation can be obtained via contrastive learning, taking advantage of the stationarity of auditory textures to …

Self-Supervised Multi-Class Pre-Training For Unsupervised Anomaly Detection And Segmentation In Medical Images, Yu Tian, Fengbei Liu, Guansong Pang, Yuanhong Chen, Yuyuan Liu, Johan W. Verjans, Rajvinder Singh

Research Collection School Of Computing and Information Systems

Unsupervised anomaly detection (UAD) that requires only normal (healthy) training images is an important tool for enabling the development of medical image analysis (MIA) applications, such as disease screening, since it is often difficult to collect and annotate abnormal (or disease) images in MIA. However, heavily relying on the normal images may cause the model training to overfit the normal class. Self-supervised pre-training is an effective solution to this problem. Unfortunately, current self-supervision methods adapted from computer vision are sub-optimal for MIA applications because they do not explore MIA domain knowledge for designing the pretext tasks or the training process. …

Conquer: Contextual Query-Aware Ranking For Video Corpus Moment Retrieval, Zhijian Hou, Chong-Wah Ngo, W. K. Chan

Research Collection School Of Computing and Information Systems

This paper tackles a recently proposed Video Corpus Moment Retrieval task. This task is essential because advanced video retrieval applications should enable users to retrieve a precise moment from a large video corpus. We propose a novel CONtextual QUery-awarE Ranking (CONQUER) model for effective moment localization and ranking. CONQUER explores query context for multi-modal fusion and representation learning in two different steps. The first step derives fusion weights for the adaptive combination of multi-modal video content. The second step performs bi-directional attention to tightly couple video and query as a single joint representation for moment localization. As query context is …

Constrained Contrastive Distribution Learning For Unsupervised Anomaly Detection And Localisation In Medical Images, Yu Tian, Guansong Pang, Fengbei Liu, Yuanhong Chen, Seon Ho Shin, Johan W. Verjans, Rajvinder Singh

Research Collection School Of Computing and Information Systems

Unsupervised anomaly detection (UAD) learns one-class classifiers exclusively with normal (i.e., healthy) images to detect any abnormal (i.e., unhealthy) samples that do not conform to the expected normal patterns. UAD has two main advantages over its fully supervised counterpart. Firstly, it is able to directly leverage large datasets available from health screening programs that contain mostly normal image samples, avoiding the costly manual labelling of abnormal samples and the subsequent issues involved in training with extremely class-imbalanced data. Further, UAD approaches can potentially detect and localise any type of lesions that deviate from the normal patterns. One significant challenge faced …

Learning Interpretable Concept Groups In Cnns, Saurabh Varshneya, Antoine Ledent, Rob Vandermeulen, Yunwen Lei, Matthias Enders, Damian Borth, Marius Kloft

Research Collection School Of Computing and Information Systems

We propose a novel training methodology---Concept Group Learning (CGL)---that encourages training of interpretable CNN filters by partitioning filters in each layer into concept groups, each of which is trained to learn a single visual concept. We achieve this through a novel regularization strategy that forces filters in the same group to be active in similar image regions for a given layer. We additionally use a regularizer to encourage a sparse weighting of the concept groups in each layer so that a few concept groups can have greater importance than others. We quantitatively evaluate CGL's model interpretability using standard interpretability evaluation …

Fine-Grained Analysis Of Structured Output Prediction, Waleed Mustafa, Yunwen Lei, Antoine Ledent, Marius And Kloft

Research Collection School Of Computing and Information Systems

In machine learning we often encounter structured output prediction problems (SOPPs), i.e. problems where the output space admits a rich internal structure. Application domains where SOPPs naturally occur include natural language processing, speech recognition, and computer vision. Typical SOPPs have an extremely large label set, which grows exponentially as a function of the size of the output. Existing generalization analysis implies generalization bounds with at least a square-root dependency on the cardinality d of the label set, which can be vacuous in practice. In this paper, we significantly improve the state of the art by developing novel high-probability bounds with …

Computational Frameworks For Multi-Robot Cooperative 3d Printing And Planning, Laxmi Prasad Poudel

Graduate Theses and Dissertations

This dissertation proposes a novel cooperative 3D printing (C3DP) approach for multi-robot additive manufacturing (AM) and presents scheduling and planning strategies that enable multi-robot cooperation in the manufacturing environment. C3DP is the first step towards achieving the overarching goal of swarm manufacturing (SM). SM is a paradigm for distributed manufacturing that envisions networks of micro-factories, each of which employs thousands of mobile robots that can manufacture different products on demand. SM breaks down the complicated supply chain used to deliver a product from a large production facility from one part of the world to another. Instead, it establishes a network …

How Important Is The Train-Validation Split In Meta-Learning?, Yu Bai, Minshuo Chen, Pan Zhou, Tuo Zhao, D. Jason Lee, Sham Kakade, Huan Wang, Caiming Xiong

Research Collection School Of Computing and Information Systems

Meta-learning aims to perform fast adaptation on a new task through learning a “prior” from multiple existing tasks. A common practice in meta-learning is to perform a train-validation split (train-val method) where the prior adapts to the task on one split of the data, and the resulting predictor is evaluated on another split. Despite its prevalence, the importance of the train-validation split is not well understood either in theory or in practice, particularly in comparison to the more direct train-train method, which uses all the pertask data for both training and evaluation. We provide a detailed theoretical study on whether …

Design And Development Of Techniques To Ensure Integrity In Fog Computing Based Databases, Abdulwahab Fahad S. Alazeb

Graduate Theses and Dissertations

The advancement of information technology in coming years will bring significant changes to the way sensitive data is processed. But the volume of generated data is rapidly growing worldwide. Technologies such as cloud computing, fog computing, and the Internet of things (IoT) will offer business service providers and consumers opportunities to obtain effective and efficient services as well as enhance their experiences and services; increased availability and higher-quality services via real-time data processing augment the potential for technology to add value to everyday experiences. This improves human life quality and easiness. As promising as these technological innovations, they are prone …

When Program Analysis Meets Bytecode Search: Targeted And Efficient Inter-Procedural Analysis Of Modern Android Apps In Backdroid, Daoyuan Wu, Debin Gao, Robert H. Deng, Rocky Chang

Research Collection School Of Computing and Information Systems

Widely-used Android static program analysis tools,e.g., Amandroid and FlowDroid, perform the whole-app interprocedural analysis that is comprehensive but fundamentallydifficult to handle modern (large) apps. The average app size hasincreased three to four times over five years. In this paper, weexplore a new paradigm of targeted inter-procedural analysis thatcan skip irrelevant code and focus only on the flows of securitysensitive sink APIs. To this end, we propose a technique calledon-the-fly bytecode search, which searches the disassembled appbytecode text just in time when a caller needs to be located. In thisway, it guides targeted (and backward) inter-procedural analysisstep by step until reaching …

Engaging Drivers Via Competition: A Case Study With Arena, Hao Cheng, Shuyu Wei, Lingyu Zhang, Zimu Zhou, Yongxin. Tong

Research Collection School Of Computing and Information Systems

Sustained work enthusiasms of drivers are crucial for the success of large-scale ride-hailing platforms. In this paper, we conduct the first-of-its-kind exploration to encourage active participation of drivers via competition. We design Arena, a competition where drivers compete for prizes via completing more trips. Through a pilot study covering over 2,600 participants, we uncover the easy-win problem, an overlooked and serious issue in competition design for real-world drivers. It refers to situations where one competitor does not show up during competition whereas the other easily wins. To solve the easy-win problem without impairing motivation of drivers, we devise a novel …

Projecting Your View Attentively: Monocular Road Scene Layout Estimation Via Cross-View Transformation, Weixiang Yang, Qi Li, Wenxi Liu, Yuanlong Yu, Yuexin Ma, Shengfeng He, Jia Pan

Research Collection School Of Computing and Information Systems

HD map reconstruction is crucial for autonomous driving. LiDAR-based methods are limited due to the deployed expensive sensors and time-consuming computation. Camera-based methods usually need to separately perform road segmentation and view transformation, which often causes distortion and the absence of content. To push the limits of the technology, we present a novel framework that enables reconstructing a local map formed by road layout and vehicle occupancy in the bird's-eye view given a front-view monocular image only. In particular, we propose a cross-view transformation module, which takes the constraint of cycle consistency between views into account and makes full use …

Learning Contextual Causality Between Daily Events From Time-Consecutive Images, Hongming Zhang, Yintong Huo, Xinran Zhao, Yangqiu Song, Dan Roth

Research Collection School Of Computing and Information Systems

Conventional textual-based causal knowledge acquisition methods typically require laborious and expensive human annotations. As a result, their scale is often limited. Moreover, as no context is provided during the annotation, the resulting causal knowledge records (e.g., ConceptNet) typically do not consider the context. In this paper, we move out of the textual domain to explore a more scalable way of acquiring causal knowledge and investigate the possibility of learning contextual causality from the visual signal. Specifically, we first propose a high-quality dataset Vis-Causal and then conduct experiments to demonstrate that with good language and visual representations, it is possible to …

A Hybrid Gaze Pointer With Voice Control, Indhuja Ravi

Master's Projects

Accessibility in technology has been a challenge since the beginning of the 1800s. Starting with building typewriters for the blind by Pellegrino Turri to the on-screen keyboard built by Microsoft, there have been several advancements towards assistive technologies. The basic tools necessary for anyone to operate a computer are to be able to navigate the device, input information, and perceive the output. All these three categories have been undergoing tremendous advancements over the years. Especially, with the internet boom, it has now become a necessity to point onto a computer screen. This has somewhat attracted research into this particular area. …

Learning Intermediate Representations For Question Answering Systems, Zakery T. Clarke

Computer Science ETDs

Question answering systems are models that can perform natural language processing (NLP) on a question, retrieve an answer from a datasource, and communicate it to a user. In question answering systems, it is important for the system to learn an underlying representation for a piece of text. There are many systems that have achieved incredible accuracy on question answering datasets such as the Stanford Question and Answer Dataset (SQuAD), but these systems often encode their knowledge in a manner that is impossible to verify. Many current models would benefit more from verifiability, than marginal accuracy improvements.

We propose a method …

City Goers: An Exploration Into Creating Seemingly Intelligent A.I. Systems, Matthew Brooke

Computer Science and Computer Engineering Undergraduate Honors Theses

Artificial Intelligence systems have come a long way over the years. One particular application of A.I. is its incorporation in video games. A key goal of creating an A.I. system in a video game is to convey a level of intellect to the player. During playtests for Halo: Combat Evolved, the developers at Bungie noticed that players deemed tougher enemies as more intelligent than weaker ones, despite the fact that there were no differences in behavior in the enemies. The tougher enemies provided a greater illusion of intelligence to the players. Inspired by this, I set out to create a …

Edgeduet: Tiling Small Object Detection For Edge Assisted Autonomous Mobile Vision, Xu Wang, Zheng Yang, Jiahang Wu, Yi Zhao, Zimu Zhou

Research Collection School Of Computing and Information Systems

Accurate, real-time object detection on resource-constrained devices enables autonomous mobile vision applications such as traffic surveillance, situational awareness, and safety inspection, where it is crucial to detect both small and large objects in crowded scenes. Prior studies either perform object detection locally on-board or offload the task to the edge/cloud. Local object detection yields low accuracy on small objects since it operates on low-resolution videos to fit in mobile memory. Offloaded object detection incurs high latency due to uploading high-resolution videos to the edge/cloud. Rather than either pure local processing or offloading, we propose to detect large objects locally while …