Digital Commons Network^™

Pa2blo: Low-Power, Personalized Audio Badge, Hemanth Sabbella, Dulaj Sanjaya Weerakoon, Manoj Gulati, Archan Misra

Research Collection School Of Computing and Information Systems

We present the hardware design and software pipeline for an ultra-low power device, in the form factor of a wearable badge, that supports energy efficient sensing, processing and wireless transfer of human voice commands and interactions. The proposed system, called PA2BLO, is envisioned to support both: (a) real-time, scalable, authorized voice based interaction and control of devices and appliances, and (b) longitudinal, low-power logging of natural voice interactions. PA2BLO in-troduces two key novel capabilities. First, it includes a low power, low-complexity voice authentication module that is able to reliably authenticate an authorized user only using low sampling rate (500 Hz) …

Demystifying Faulty Code: Step-By-Step Reasoning For Explainable Fault Localization, Ratnadira Widyasari, Jia Wei Ang, Truong Giang Nguyen, Neil Sharma, David Lo

Research Collection School Of Computing and Information Systems

Fault localization is a critical process that involves identifying specific program elements responsible for program failures. Manually pinpointing these elements, such as classes, methods, or statements, which are associated with a fault is laborious and time-consuming. To overcome this challenge, various fault localization tools have been developed. These tools typically generate a ranked list of suspicious program elements. However, this information alone is insufficient. A prior study emphasized that automated fault localization should offer a rationale. In this study, we investigate the step-by-step reasoning for explainable fault localization. We explore the potential of Large Language Models (LLM) in assisting developers …

Screening Through A Broad Pool: Towards Better Diversity For Lexically Constrained Text Generation, Changsen Yuan, Heyan Huang, Yixin Cao, Qianwen Cao

Research Collection School Of Computing and Information Systems

Lexically constrained text generation (CTG) is to generate text that contains given constrained keywords. However, the text diversity of existing models is still unsatisfactory. In this paper, we propose a lightweight dynamic refinement strategy that aims at increasing the randomness of inference to improve generation richness and diversity while maintaining a high level of fluidity and integrity. Our basic idea is to enlarge the number and length of candidate sentences in each iteration, and choose the best for subsequent refinement. On the one hand, different from previous works, which carefully insert one token between two words per action, we insert …

Weakly Supervised Attention-Based Recognition Under Spectral, Turbulence, And Resource Variations, Kshitij Naresh Nikhal

Dissertations and Doctoral Documents, University of Nebraska-Lincoln, 2023–

While supervised optimization paradigms are ubiquitous across diverse recognition systems, the risk of over-fitting and increasing bias have limited their applicability.

This dissertation focuses on unsupervised learning—learning without precisely curated data—and argues that unsupervised learning methods can enable both discriminability and generalizability. Through the use of attention-based machine learning and advanced clustering, unsupervised methods are able to focus on fine-grained information in images without any explicit supervision. The dissertation introduces a domain-bridging framework for tasks like cross-spectrum matching and long-range recognition, utilizing intra-domain clustering and inter-domain matching to generate pseudo-labels. Additionally, a hash-based network is proposed to accelerate the search …

Continual Online Learning-Based Optimal Tracking Control Of Nonlinear Strict-Feedback Systems: Application To Unmanned Aerial Vehicles, Irfan Ganie, Sarangapani Jagannathan

Electrical and Computer Engineering Faculty Research & Creative Works

A novel optimal trajectory tracking scheme is introduced for nonlinear continuous-time systems in strict feedback form with uncertain dynamics by using neural networks (NNs). The method employs an actor-critic-based NN back-stepping technique for minimizing a discounted value function along with an identifier to approximate unknown system dynamics that are expressed in augmented form. Novel online weight update laws for the actor and critic NNs are derived by using both the NN identifier and Hamilton-Jacobi-Bellman residual error. A new continual lifelong learning technique utilizing the Fisher Information Matrix via Hamilton-Jacobi-Bellman residual error is introduced to obtain the significance of weights in …

Mask2former With Improved Query For Semantic Segmentation In Remote-Sensing Images, Shichen Guo, Qi Wang, Shiming Xiang, Shuwen Wang, Xuezhi Wang

Computer Science Faculty Publications and Presentations

Semantic segmentation of remote sensing (RS) images is vital in various practical applications, including urban construction planning, natural disaster monitoring, and land resources investigation. However, RS images are captured by airplanes or satellites at high altitudes and long distances, resulting in ground objects of the same category being scattered in various corners of the image. Moreover, objects of different sizes appear simultaneously in RS images. For example, some objects occupy a large area in urban scenes, while others only have small regions. Technically, the above two universal situations pose significant challenges to the segmentation with a high quality for RS …

Voice Synthesis Improvement By Machine Learning Of Natural Prosody, Joseph Kane, Michael N. Johnstone, Patryk Szewczyk

Research outputs 2022 to 2026

Since the advent of modern computing, researchers have striven to make the human–computer interface (HCI) as seamless as possible. Progress has been made on various fronts, e.g., the desktop metaphor (interface design) and natural language processing (input). One area receiving attention recently is voice activation and its corollary, computer-generated speech. Despite decades of research and development, most computer-generated voices remain easily identifiable as non-human. Prosody in speech has two primary components—intonation and rhythm—both often lacking in computer-generated voices. This research aims to enhance computer-generated text-to-speech algorithms by incorporating melodic and prosodic elements of human speech. This study explores a novel …

Fuzzy Software Reliability And Optimal Release Policy With Log-Logistic Testing Effort: An Analysis, Seema Rani, Jitendra Kumar, N. Ahmad

Applications and Applied Mathematics: An International Journal (AAM)

We will discuss a Software Reliability Growth Model (SRGM) using fuzzy and imperfect debugging environments; we integrate Log-Logistic (LL) Testing Effort Function (TEF) into fuzzy SRGMs. Estimation methods, such as Least Square and Maximum Likelihood, are used to obtain the value of Testing-Effort and SRGMs parameters. It is not always possible and is constantly required to quantify the exact value of parameters. Due to human conduct, the value of Testing-Effort and SRGM parameters cannot be exactly quantified. In this scenario, parameters are supposed to be vague or fuzzy. To make the software consistent, the developer needs to propose some quantity …

Eyris: From The Lab To The Market, Steven Miller, David Gomulya, Mahima Rao-Kachroo

Asian Management Insights

Singapore’s trailblazer AI algorithm for detecting diabetes-related eye diseases. Can you imagine getting the results of your eye disease screening within minutes rather than days? This capability is what EyRIS, a Singapore-based start-up that uses the AI (Artificial Intelligence)-driven Singapore Eye LEsion Analyzer (SELENA+) algorithm to screen for diabetes-related eye diseases, set out to productise and commercialise.

Superminds At Work: The Promise Of Human-Ai Collaboration, Thomas W. Malone

Asian Management Insights

Massachusetts Institute of Technology (MIT) Center for Collective Intelligence Director Professor Thomas W. Malone’s scholarship offers deep insights into the promise afforded by the synergies between human intelligence and technology. According to Professor Malone, the boundaries between human intellect and technological prowess are becoming increasingly blurred, but this may not be a bad thing for humankind. In Asian Management Insights’ inaugural Pulse Point interview, we get to learn more about the concept of ‘collective intelligence’, which explores how a partnership between humans and Artificial Intelligence (AI) can be catalysed to make ground-breaking advancements in addressing the wicked problems of our …

Hello, Jarvis, Archan Misra

Asian Management Insights

How AI-enabled interactive agents will reshape our workforce of today and tomorrow.

Stability Of Predator-Prey Model For Worm Attack In Wireless Sensor Networks, Rajeev Kishore, Padam Singh

Applications and Applied Mathematics: An International Journal (AAM)

In this paper, we propose a predator-prey mathematical model for analyzing the dynamical behaviors of the system. This system is an epidemic model, and it is capable of ascertaining the worm's spreading at the initial stage and improving the security of wireless sensor networks. We investigate different fixed points and examine the stability of the projected model.

Developing Singapore As A Smart Nation, Josephine Teo

Asian Management Insights

Mrs Josephine Teo, Singapore’s Minister for Communications and Information, and Minister-in-charge of Smart Nation and Cybersecurity, speaks about leading the country’s Smart Nation drive.

Technical Note: Generalizable And Promptable Artificial Intelligence Model To Augment Clinical Delineation In Radiation Oncology, Lian Zhang, Zhengliang Liu, Lu Zhang, Zihao Wu, Xiaowei Yu, Jason Holmes, Hongying Feng, Haixing Dai, Xiang Li, Quanzheng Li, William W. Wong, Sujay A. Vora, Dajiang Zhu, Tianming Liu, Wei Liu

Computer Science Faculty Research & Creative Works

Background: Efficient and accurate delineation of organs at risk (OARs) is a critical procedure for treatment planning and dose evaluation. Deep learning-based auto-segmentation of OARs has shown promising results and is increasingly being used in radiation therapy. However, existing deep learning-based auto-segmentation approaches face two challenges in clinical practice: generalizability and human-AI interaction. A generalizable and prompt able auto-segmentation model, which segments OARs of multiple disease sites simultaneously and supports on-the-fly human-AI interaction, can significantly enhance the efficiency of radiation therapy treatment planning. Purpose: Meta's segment anything model (SAM) was proposed as a generalizable and prompt able model for next-generation …

Test-Time Augmentation For 3d Point Cloud Classification And Segmentation, Tuan-Anh Vu, Srinjay Sarkar, Zhiyuan Zhang, Binh-Son Hua, Sai-Kit Yeung

Research Collection School Of Computing and Information Systems

Data augmentation is a powerful technique to enhance the performance of a deep learning task but has received less attention in 3D deep learning. It is well known that when 3D shapes are sparsely represented with low point density, the performance of the downstream tasks drops significantly. This work explores test-time augmentation (TTA) for 3D point clouds. We are inspired by the recent revolution of learning implicit representation and point cloud upsampling, which can produce high-quality 3D surface reconstruction and proximity-to-surface, respectively. Our idea is to leverage the implicit field reconstruction or point cloud upsampling techniques as a systematic way …

Iterative Graph Self-Distillation, Hanlin Zhang, Shuai Lin, Weiyang Liu, Pan Zhou, Jian Tang, Xiaodan Liang, Eric Xing

Research Collection School Of Computing and Information Systems

Recently, there has been increasing interest in the challenge of how to discriminatively vectorize graphs. To address this, we propose a method called Iterative Graph Self-Distillation (IGSD) which learns graph-level representation in an unsupervised manner through instance discrimination using a self-supervised contrastive learning approach. IGSD involves a teacher-student distillation process that uses graph diffusion augmentations and constructs the teacher model using an exponential moving average of the student model. The intuition behind IGSD is to predict the teacher network representation of the graph pairs under different augmented views. As a natural extension, we also apply IGSD to semi-supervised scenarios by …

Towards Understanding Convergence And Generalization Of Adamw, Pan Zhou, Xingyu Xie, Zhouchen Lin, Shuicheng Yan

Research Collection School Of Computing and Information Systems

AdamW modifies Adam by adding a decoupled weight decay to decay network weights per training iteration. For adaptive algorithms, this decoupled weight decay does not affect specific optimization steps, and differs from the widely used ℓ2-regularizer which changes optimization steps via changing the first- and second-order gradient moments. Despite its great practical success, for AdamW, its convergence behavior and generalization improvement over Adam and ℓ2-regularized Adam (ℓ2-Adam) remain absent yet. To solve this issue, we prove the convergence of AdamW and justify its generalization advantages over Adam and ℓ2-Adam. Specifically, AdamW provably converges but minimizes a dynamically regularized loss that …

Stability Verification In Stochastic Control Systems Via Neural Network Supermartingales, Mathias Lechner, Dorde Zikelic, Krishnendu Chatterjee, Thomas A. Henzinger

Research Collection School Of Computing and Information Systems

We consider the problem of formally verifying almost-sure (a.s.) asymptotic stability in discrete-time nonlinear stochastic control systems. While verifying stability in deterministic control systems is extensively studied in the literature, verifying stability in stochastic control systems is an open problem. The few existing works on this topic either consider only specialized forms of stochasticity or make restrictive assumptions on the system, rendering them inapplicable to learning algorithms with neural network policies. In this work, we present an approach for general nonlinear stochastic control problems with two novel aspects: (a) instead of classical stochastic extensions of Lyapunov functions, we use ranking …

Self-Admitted Technical Debts Identification: How Far Are We?, Hao Gu, Shichao Zhang, Qiao Huang, Zhifang Liao, Jiakun Liu, David Lo

Research Collection School Of Computing and Information Systems

Self-admitted technical debt (SATD) is a kind of technical debt that is already acknowledged by the developers and needs additional work or resources to address in the future. In recent years, though many methods have been proposed to detect SATDs, these methods have mainly focused on Java-type code comments published by Maldonado et al. It is unclear whether these methods trained on Maldonado's code comments dataset can find SATD in other programming languages or other software artifacts, such as issue trackers, pull requests, and commit messages effectively. In order to answer the above confusion and investigate how far our community …

Sustainability Forecasting For Deep Learning Packages, Junxiao Han, Yunkun Wang, Zhongxin Liu, Lingfeng Bao, Jiakun Liu, David Lo, Shuiguang Deng

Research Collection School Of Computing and Information Systems

Deep Learning (DL) technologies have been widely adopted to tackle various tasks. In this process, through software dependencies, a multi-layer DL supply chain (SC) is formed, with DL frameworks acting as the root, DL packages acting as the bridge nodes, and downstream DL projects acting as the periphery. However, most Open Source Software (OSS) projects may fail. Considering the crucial position of DL packages in the DL SC, to foster the sustainable development of DL SCs and DL packages, we aim to forecast the long-term sustainability of DL packages. Here, sustained activity is adopted as the main proxy of sustainability, …

Gsword: Gpu-Accelerated Sampling For Subgraph Counting, Chang Ye, Yuchen Li, Shixuan Sun, Wentian Guo

Research Collection School Of Computing and Information Systems

Subgraph counting is a fundamental component for many downstream applications such as graph representation learning and query optimization. Since obtaining the exact count is often intractable, there have been a plethora of approximation methods on graph sampling techniques. Nonetheless, the state-of-the-art sampling methods still require massive samples to produce accurate approximations on large data graphs. We propose gSWORD, a GPU framework that leverages the massive parallelism of GPUs to accelerate iterative sampling algorithms for subgraph counting. Despite the embarrassingly parallel nature of the samples, there are unique challenges in accelerating subgraph counting due to its irregular computation logic. To address …

Improvement Of Flow Coefficient Estimation With Limited Well Test Data For Real-Time Condition Analytics Of Choke Valve, T. Visawameteekul, Tin Seong Kam, P. Thamvechvitee

Research Collection School Of Computing and Information Systems

The study discusses a method for monitoring the internal condition of choke valves to predict sand erosion using flow coefficient (Cv). The method calculates the Cv of the choke valve by utilizing eight parameters and compares it to the newly manufactured value to generate warnings. However, the availability of spot-check well test data can significantly impact the model's efficiency if tests are performed infrequently. To address this issue, the Extended Cv monitoring method is proposed in this paper. The main purpose of this study is to develop a model for estimating Cv value in the absence of well test data. …

Transiam: Aggregating Multi-Modal Visual Features With Locality For Medical Image Segmentation, Xuejian Li, Shiqiang Ma, Junhai Xu, Jijun Tang, Shengfeng He, Fei Guo

Research Collection School Of Computing and Information Systems

Automatic segmentation of medical images plays an important role in the diagnosis of diseases. On single-modal data, convolutional neural networks have demonstrated satisfactory performance. However, multi-modal data encompasses a greater amount of information rather than single-modal data. Multi-modal data can be effectively used to improve the segmentation accuracy of regions of interest by analyzing both spatial and temporal information. In this study, we propose a dual-path segmentation model for multi-modal medical images, named TranSiam. Taking into account that there is a significant diversity between the different modalities, TranSiam employs two parallel CNNs to extract the features which are specific to …

Simulated Annealing With Reinforcement Learning For The Set Team Orienteering Problem With Time Windows, Vincent F. Yu, Nabila Y. Salsabila, Shih-W Lin, Aldy Gunawan

Research Collection School Of Computing and Information Systems

This research investigates the Set Team Orienteering Problem with Time Windows (STOPTW), a new variant of the well-known Team Orienteering Problem with Time Windows and Set Orienteering Problem. In the STOPTW, customers are grouped into clusters. Each cluster is associated with a profit attainable when a customer in the cluster is visited within the customer's time window. A Mixed Integer Linear Programming model is formulated for STOPTW to maximizing total profit while adhering to time window constraints. Since STOPTW is an NP-hard problem, a Simulated Annealing with Reinforcement Learning (SARL) algorithm is developed. The proposed SARL incorporates the core concepts …

Non-Monotonic Generation Of Knowledge Paths For Context Understanding, Pei-Chi Lo, Ee-Peng Lim

Research Collection School Of Computing and Information Systems

Knowledge graphs can be used to enhance text search and access by augmenting textual content with relevant background knowledge. While many large knowledge graphs are available, using them to make semantic connections between entities mentioned in the textual content remains to be a difficult task. In this work, we therefore introduce contextual path generation (CPG) which refers to the task of generating knowledge paths, contextual path, to explain the semantic connections between entities mentioned in textual documents with given knowledge graph. To perform CPG task well, one has to address its three challenges, namely path relevance, incomplete knowledge graph, and …

Non-Binary Evaluation Of Next-Basket Food Recommendation, Yue Liu, Palakorn Achananuparp, Ee-Peng Lim

Research Collection School Of Computing and Information Systems

Next-basket recommendation (NBR) is a recommendation task that predicts a basket or a set of items a user is likely to adopt next based on his/her history of basket adoption sequences. It enables a wide range of novel applications and services from predicting next basket of items for grocery shopping to recommending food items a user is likely to consume together in the next meal. Even though much progress has been made in the algorithmic NBR research over the years, little research has been done to broaden knowledge about the evaluation of NBR methods, which is largely based on the …

Analyzing Biomedical Datasets With Symbolic Tree Adaptive Resonance Theory, Sasha Petrenko, Daniel B. Hier, Mary A. Bone, Tayo Obafemi-Ajayi, Erik J. Timpson, William E. Marsh, Michael Speight, Donald C. Wunsch

Chemistry Faculty Research & Creative Works

Biomedical Datasets Distill Many Mechanisms Of Human Diseases, Linking Diseases To Genes And Phenotypes (Signs And Symptoms Of Disease), Genetic Mutations To Altered Protein Structures, And Altered Proteins To Changes In Molecular Functions And Biological Processes. It Is Desirable To Gain New Insights From These Data, Especially With Regard To The Uncovering Of Hierarchical Structures Relating Disease Variants. However, Analysis To This End Has Proven Difficult Due To The Complexity Of The Connections Between Multi-Categorical Symbolic Data. This Article Proposes Symbolic Tree Adaptive Resonance Theory (START), With Additional Supervised, Dual-Vigilance (DV-START), And Distributed Dual-Vigilance (DDV-START) Formulations, For The Clustering Of …

Fixing Your Own Smells: Adding A Mistake-Based Familiarization Step When Teaching Code Refactoring, Ivan Wei Han Tan, Christopher M. Poskitt

Research Collection School Of Computing and Information Systems

Programming problems can be solved in a multitude of functionally correct ways, but the quality of these solutions (e.g. readability, maintainability) can vary immensely. When code quality is poor, symptoms emerge in the form of 'code smells', which are specific negative characteristics (e.g. duplicate code) that can be resolved by applying refactoring patterns. Many undergraduate computing curricula train students on this software engineering practice, often doing so via exercises on unfamiliar instructor-provided code. Our observation, however, is that this makes it harder for novices to internalise refactoring as part of their own development practices. In this paper, we propose a …

Ditmos: Delving Into Diverse Tiny-Model Selection On Microcontrollers, Xiao Ma, Shengfeng He, Hezhe Qiao, Dong Ma

Research Collection School Of Computing and Information Systems

Enabling efficient and accurate deep neural network (DNN) inference on microcontrollers is non-trivial due to the constrained on-chip resources. Current methodologies primarily focus on compressing larger models yet at the expense of model accuracy. In this paper, we rethink the problem from the inverse perspective by constructing small/weak models directly and improving their accuracy. Thus, we introduce DiTMoS, a novel DNN training and inference framework with a selectorclassifiers architecture, where the selector routes each input sample to the appropriate classifier for classification. DiTMoS is grounded on a key insight: a composition of weak models can exhibit high diversity and the …

Active Discovering New Slots For Task-Oriented Conversation, Yuxia Wu, Tianhao Dai, Zhedong Zheng, Lizi Liao

Research Collection School Of Computing and Information Systems

Existing task-oriented conversational systems heavily rely on domain ontologies with pre-defined slots and candidate values. In practical settings, these prerequisites are hard to meet, due to the emerging new user requirements and ever-changing scenarios. To mitigate these issues for better interaction performance, there are efforts working towards detecting out-of-vocabulary values or discovering new slots under unsupervised or semi-supervised learning paradigms. However, overemphasizing on the conversation data patterns alone induces these methods to yield noisy and arbitrary slot results. To facilitate the pragmatic utility, real-world systems tend to provide a stringent amount of human labeling quota, which offers an authoritative way …