Vitus J. Leung
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Vitus J. Leung has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Ayse K. Coskun
Boston University
7 shared publications - Manuel Egele
Boston University, Boston, USA
5 shared publications - Benjamin Schwaller
Sandia National Laboratories, Albuquerque, NM, USA
4 shared publications - Brian Kulis
Boston University
4 shared publications - Burak Aksar
Boston University
3 shared publications - Efe Sencan
Boston University, Boston, MA, USA
3 shared publications
Latest Publications
- 2 January 2026
- Published by Springer Nature
- p. 21-40
Job Grouping Based Intelligent Resource Prediction Framework
- 2 January 2026
- Published by Springer Nature
- p. 21-40
Job Grouping Based Intelligent Resource Prediction Framework
- Open Access
UniCoMTE: A Universal Counterfactual Framework for Explaining Time-Series Classifiers on ECG Data
Machine learning models, particularly deep neural networks, have demonstrated strong performance in classifying complex time series data. However, their black-box nature limits trust and adoption, especially in high-stakes domains such as healthcare. To address this challenge, we introduce UniCoMTE, a model-agnostic framework for generating counterfactual explanations for multivariate time series classifiers. The framework identifies temporal features that most heavily influence a model's prediction by modifying the input sample and assessing its impact on the model's prediction. UniCoMTE is compatible with a wide range of model architectures and operates directly on raw time series inputs. In this study, we evaluate UniCoMTE's explanations on a time series ECG classifier. We quantify explanation quality by comparing our explanations' comprehensibility to comprehensibility of established techniques (LIME and SHAP) and assessing their generalizability to similar samples. Furthermore, clinical utility is assessed through a questionnaire completed by medical experts who review counterfactual explanations presented alongside original ECG samples. Results show that our approach produces concise, stable, and human-aligned explanations that outperform existing methods in both clarity and applicability. By linking model predictions to meaningful signal patterns, the framework advances the interpretability of deep learning models for real-world time series applications.
Open AccessUniCoMTE: A Universal Counterfactual Framework for Explaining Time-Series Classifiers on ECG Data
Machine learning models, particularly deep neural networks, have demonstrated strong performance in classifying complex time series data. However, their black-box nature limits trust and adoption, especially in high-stakes domains such as healthcare. To address this challenge, we introduce UniCoMTE, a model-agnostic framework for generating counterfactual explanations for multivariate time series classifiers. The framework identifies temporal features that most heavily influence a model's prediction by modifying the input sample and assessing its impact on the model's prediction. UniCoMTE is compatible with a wide range of model architectures and operates directly on raw time series inputs. In this study, we evaluate UniCoMTE's explanations on a time series ECG classifier. We quantify explanation quality by comparing our explanations' comprehensibility to comprehensibility of established techniques (LIME and SHAP) and assessing their generalizability to similar samples. Furthermore, clinical utility is assessed through a questionnaire completed by medical experts who review counterfactual explanations presented alongside original ECG samples. Results show that our approach produces concise, stable, and human-aligned explanations that outperform existing methods in both clarity and applicability. By linking model predictions to meaningful signal patterns, the framework advances the interpretability of deep learning models for real-world time series applications.
- 13 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 35 (4) , 693-706
Runtime Performance Anomaly Diagnosis in Production HPC Systems Using Active Learning
With the increasing scale and complexity of High-Performance Computing (HPC) systems, performance variations in applications caused by anomalies have become significant bottlenecks in system health and operational efficiency. As we move towards exascale systems, these variations become more prominent due to the increased sharing of resources. Such variations lead to lower energy efficiency and higher operational costs. To mitigate these problems, one must quickly and accurately diagnose the root cause of the anomalies at scale. One way to evaluate system health and identify the underlying causes is by manually examining certain performance metrics in telemetry data or using rule-based methods. Due to the daily size of telemetry data reaching terabytes and the fact that the numeric telemetry data contains thousands of metrics, manual analysis of telemetry to diagnose problems becomes challenging. Given these limitations, Machine Learning (ML)-based approaches have been gaining popularity as they have been shown to be effective and practical in diagnosing previously encountered performance anomalies. One primary challenge for supervised ML models is that they require a significant amount of labeled samples during training. However, obtaining many labels for anomalies is extremely difficult and costly, considering anomalies occur infrequently and real-world numeric system telemetry data is hard to label since it contains thousands of metrics. This paper proposes a novel active learning-based framework that diagnoses performance anomalies (i.e., identifying the type of an anomaly) in HPC systems at runtime using significantly fewer labeled samples compared to state-of-the-art ML-based approaches. We show that the proposed framework achieves the same F1-score compared to a supervised approach using much fewer labeled samples (i.e., 16x fewer samples for achieving a 0.78 F1-score, 11x fewer samples for achieving a 0.82 F1-score), even when there are previously unseen applications and application inputs in the test dataset.
- 13 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 35 (4) , 693-706
Runtime Performance Anomaly Diagnosis in Production HPC Systems Using Active Learning
With the increasing scale and complexity of High-Performance Computing (HPC) systems, performance variations in applications caused by anomalies have become significant bottlenecks in system health and operational efficiency. As we move towards exascale systems, these variations become more prominent due to the increased sharing of resources. Such variations lead to lower energy efficiency and higher operational costs. To mitigate these problems, one must quickly and accurately diagnose the root cause of the anomalies at scale. One way to evaluate system health and identify the underlying causes is by manually examining certain performance metrics in telemetry data or using rule-based methods. Due to the daily size of telemetry data reaching terabytes and the fact that the numeric telemetry data contains thousands of metrics, manual analysis of telemetry to diagnose problems becomes challenging. Given these limitations, Machine Learning (ML)-based approaches have been gaining popularity as they have been shown to be effective and practical in diagnosing previously encountered performance anomalies. One primary challenge for supervised ML models is that they require a significant amount of labeled samples during training. However, obtaining many labels for anomalies is extremely difficult and costly, considering anomalies occur infrequently and real-world numeric system telemetry data is hard to label since it contains thousands of metrics. This paper proposes a novel active learning-based framework that diagnoses performance anomalies (i.e., identifying the type of an anomaly) in HPC systems at runtime using significantly fewer labeled samples compared to state-of-the-art ML-based approaches. We show that the proposed framework achieves the same F1-score compared to a supervised approach using much fewer labeled samples (i.e., 16x fewer samples for achieving a 0.78 F1-score, 11x fewer samples for achieving a 0.82 F1-score), even when there are previously unseen applications and application inputs in the test dataset.
- Open Access
- 11 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-14
Prodigy: Towards Unsupervised Anomaly Detection in Production HPC Systems
Performance variations caused by anomalies in modern High Performance Computing (HPC) systems lead to decreased efficiency, impaired application performance, and increased operational costs. While machine learning (ML)-based frameworks for automated anomaly detection (often based on time series telemetry data) are gaining popularity in the literature, practical deployment challenges are often overlooked. Some ML-based frameworks require extensive customization, while others need a rich set of labeled samples, none of which are feasible for a production HPC system. This paper introduces a variational autoencoder-based anomaly detection framework, Prodigy, that outperforms the state-of-the-art alternatives by achieving a 0.95 F1-score when detecting performance anomalies. The paper also provides a real system implementation of Prodigy that enables easy integration with monitoring frameworks and rapid deployment. We deploy Prodigy on a production HPC system and demonstrate 88% accuracy in detecting anomalies. Prodigy involves an interface to provide job- and node-level analysis and explanations for anomaly predictions.
- 11 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-14
Open AccessProdigy: Towards Unsupervised Anomaly Detection in Production HPC Systems
Performance variations caused by anomalies in modern High Performance Computing (HPC) systems lead to decreased efficiency, impaired application performance, and increased operational costs. While machine learning (ML)-based frameworks for automated anomaly detection (often based on time series telemetry data) are gaining popularity in the literature, practical deployment challenges are often overlooked. Some ML-based frameworks require extensive customization, while others need a rich set of labeled samples, none of which are feasible for a production HPC system. This paper introduces a variational autoencoder-based anomaly detection framework, Prodigy, that outperforms the state-of-the-art alternatives by achieving a 0.95 F1-score when detecting performance anomalies. The paper also provides a real system implementation of Prodigy that enables easy integration with monitoring frameworks and rapid deployment. We deploy Prodigy on a production HPC system and demonstrate 88% accuracy in detecting anomalies. Prodigy involves an interface to provide job- and node-level analysis and explanations for anomaly predictions.
- Open Access
- 10 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Towards Practical Machine Learning Frameworks for Performance Diagnostics in Supercomputers
- 10 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessTowards Practical Machine Learning Frameworks for Performance Diagnostics in Supercomputers
- 14 September 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 30 (4) , 883-896
Online Diagnosis of Performance Variation in HPC Systems Using Machine Learning
As the size and complexity of HPC systems grow in line with advancements in hardware and software technology, HPC systems increasingly suffer from performance variation due to shared resource contention as well as software- and hardware-related problems. Such performance variations can lead to failures and inefficiencies, and are among the main challenges in system resiliency. To minimize the impact of performance variation, one must quickly and accurately detect and diagnose the anomalies that cause the variation and take mitigating actions. However, it is difficult to identify anomalies based on the voluminous, high-dimensional, and noisy data collected by system monitoring infrastructures. This paper presents a novel machine learning based framework to automatically diagnose performance anomalies at runtime. Our framework leverages historical resource usage data to extract signatures of previously-observed anomalies. We first convert the collected time series data into easy-to-compute statistical features. We then identify the features that are required to detect anomalies, and extract the signatures of these anomalies. At runtime, we use these signatures to diagnose anomalies with negligible overhead. We evaluate our framework using experiments on a real-world HPC supercomputer and demonstrate that our approach successfully identifies 98% of injected anomalies and consistently outperforms existing anomaly diagnosis techniques.
- 14 September 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 30 (4) , 883-896
Online Diagnosis of Performance Variation in HPC Systems Using Machine Learning
As the size and complexity of HPC systems grow in line with advancements in hardware and software technology, HPC systems increasingly suffer from performance variation due to shared resource contention as well as software- and hardware-related problems. Such performance variations can lead to failures and inefficiencies, and are among the main challenges in system resiliency. To minimize the impact of performance variation, one must quickly and accurately detect and diagnose the anomalies that cause the variation and take mitigating actions. However, it is difficult to identify anomalies based on the voluminous, high-dimensional, and noisy data collected by system monitoring infrastructures. This paper presents a novel machine learning based framework to automatically diagnose performance anomalies at runtime. Our framework leverages historical resource usage data to extract signatures of previously-observed anomalies. We first convert the collected time series data into easy-to-compute statistical features. We then identify the features that are required to detect anomalies, and extract the signatures of these anomalies. At runtime, we use these signatures to diagnose anomalies with negligible overhead. We evaluate our framework using experiments on a real-world HPC supercomputer and demonstrate that our approach successfully identifies 98% of injected anomalies and consistently outperforms existing anomaly diagnosis techniques.
- 26 June 2014
- Published by Elsevier in Sustainable Computing: Informatics and Systems
- Vol. 6, 48-57
Simulation and optimization of HPC job allocation for jointly reducing communication and cooling costs
- 26 June 2014
- Published by Elsevier in Sustainable Computing: Informatics and Systems
- Vol. 6, 48-57
Simulation and optimization of HPC job allocation for jointly reducing communication and cooling costs
- 1 July 2009
- Published by American Society of Civil Engineers (ASCE) in Journal of Water Resources Planning and Management
- Vol. 135 (4) , 253-263
Designing Contamination Warning Systems for Municipal Water Networks Using Imperfect Sensors
We consider the problem of designing a contaminant warning system for a municipal water distribution network that uses imperfect sensors, which can generate false-positive and false-negative detections. Although sensor placement optimization methods have been developed for contaminant warning systems, most sensor placement formulations assume perfect sensors, which does not accurately reflect the behavior of real sensor technology. We describe a general exact nonlinear formulation for imperfect sensors and a linear approximation. We consider six general solution strategies, some of which have multiple solution methods. We applied these methods to three test networks, including one with over 10,000 nodes. Our experiments indicate that it is worth deploying a sensor network even when sensors have low detection probability. They also indicate it is worth paying attention to sensor imperfections when placing sensors even when there is a response delay of up to 8 h. The best choice of solution strategy depends...
- 1 July 2009
- Published by American Society of Civil Engineers (ASCE) in Journal of Water Resources Planning and Management
- Vol. 135 (4) , 253-263
Designing Contamination Warning Systems for Municipal Water Networks Using Imperfect Sensors
We consider the problem of designing a contaminant warning system for a municipal water distribution network that uses imperfect sensors, which can generate false-positive and false-negative detections. Although sensor placement optimization methods have been developed for contaminant warning systems, most sensor placement formulations assume perfect sensors, which does not accurately reflect the behavior of real sensor technology. We describe a general exact nonlinear formulation for imperfect sensors and a linear approximation. We consider six general solution strategies, some of which have multiple solution methods. We applied these methods to three test networks, including one with over 10,000 nodes. Our experiments indicate that it is worth deploying a sensor network even when sensors have low detection probability. They also indicate it is worth paying attention to sensor imperfections when placing sensors even when there is a response delay of up to 8 h. The best choice of solution strategy depends...
- 13 March 2008
- Published by American Society of Civil Engineers (ASCE)
- p. 1-13
On the Placement of Imperfect Sensors in Municipal Water Networks
We consider the problem of optimally placing water quality sensors in municipal water networks under the assumption that sensors may fail. We give a non-linear formulation of the problem, then a linearization of this formulation in the form of a mixed-integer program (MIP). We explore the scalability limits of this formulation, then use it as a bounding procedure for a local search heuristic that optimizes the same objective: minimizing the expected impact of a contamination event. This heuristic can find optimal or near-optimal solutions on networks with over ten thousand junctions. This paper was presented at the 8th Annual Water Distribution Systems Analysis Symposium which was held with the generous support of Awwa Research Foundation (AwwaRF).
- 13 March 2008
- Published by American Society of Civil Engineers (ASCE)
- p. 1-13
On the Placement of Imperfect Sensors in Municipal Water Networks
We consider the problem of optimally placing water quality sensors in municipal water networks under the assumption that sensors may fail. We give a non-linear formulation of the problem, then a linearization of this formulation in the form of a mixed-integer program (MIP). We explore the scalability limits of this formulation, then use it as a bounding procedure for a local search heuristic that optimizes the same objective: minimizing the expected impact of a contamination event. This heuristic can find optimal or near-optimal solutions on networks with over ten thousand junctions. This paper was presented at the 8th Annual Water Distribution Systems Analysis Symposium which was held with the generous support of Awwa Research Foundation (AwwaRF).
- 9 September 2007
- Published by Springer Nature in Algorithmica
- Vol. 50 (2) , 279-298
Communication-Aware Processor Allocation for Supercomputers: Finding Point Sets of Small Average Distance
- 9 September 2007
- Published by Springer Nature in Algorithmica
- Vol. 50 (2) , 279-298
Communication-Aware Processor Allocation for Supercomputers: Finding Point Sets of Small Average Distance