arXiv subjects

Vitus J. Leung

Publications and source records attributed to Vitus J. Leung.

5 recordsLinked to original sources

arXiv2025

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.

cs.LG↗

arXiv2020

Counterfactual Explanations for Machine Learning on Multivariate Time Series Data

Applying machine learning (ML) on multivariate time series data has growing popularity in many application domains, including in computer system management. For example, recent high performance computing (HPC) research proposes a variety of ML frameworks that use system telemetry data in the form of multivariate time series so as to detect performance variations, perform intelligent scheduling or node allocation, and improve system security. Common barriers for adoption for these ML frameworks include the lack of user trust and the difficulty of debugging. These barriers need to be overcome to enable the widespread adoption of ML frameworks in production systems. To address this challenge, this paper proposes a novel explainability technique for providing counterfactual explanations for supervised ML frameworks that use multivariate time series data. The proposed method outperforms state-of-the-art explainability methods on several different ML frameworks and data sets in metrics such as faithfulness and robustness. The paper also demonstrates how the proposed method can be used to debug ML frameworks and gain a better understanding of HPC system telemetry data.

cs.LG↗

arXiv2010

Sensitivity of the Performance of a Simple Exchange Model to its Topology

We study a simple exchange model in which price is fixed and the amount of a good transferred between actors depends only on the actors' respective budgets and the existence of a link between transacting actors. The model induces a simply-connected but possibly multi-component bipartite graph. A trading session on a fixed graph consists of a sequence of exchanges between connected buyers and sellers until no more exchanges are possible. We deem a trading session "feasible" if all of the buyers satisfy their respective demands. If all trading sessions are feasible the graph is deemed "successful", otherwise the feasibility of a trading session depends on the order of the sequence of exchanges. We demonstrate that topology is important for the success of trading sessions on graphs. In particular, for the case that supply equals demand for each component of the graph, we prove that the graph is successful if and only if the graph consists of components each of which are complete bipartite. For the case that supply exceeds demand, we prove that the other topologies also can be made successful but with finite reserve (i.e., excess supply) requirements that may grow proportional to the number of buyers. Finally, with computations for a small instance of the model, we provide an example of the wide range of performance in which only the connectivity varies. These results taken together place limits on the improvements in performance that can be expected from proposals to increase the connectivity of sparse exchange networks.

q-fin.TR↗

arXiv2007

Community Detection via Facility Location

In this paper we apply theoretical and practical results from facility location theory to the problem of community detection in networks. The result is an algorithm that computes bounds on a minimization variant of local modularity. We also define the concept of an edge support and a new measure of the goodness of community structures with respect to this concept. We present preliminary results and note that our methods are massively parallelizable.

physics.soc-ph↗

arXiv2005

Communication-Aware Processor Allocation for Supercomputers

This paper gives processor-allocation algorithms for minimizing the average number of communication hops between the assigned processors for grid architectures, in the presence of occupied cells. The simpler problem of assigning processors on a free grid has been studied by Karp, McKellar, and Wong who show that the solutions have nontrivial structure; they left open the complexity of the problem. The associated clustering problem is as follows: Given n points in Re^d, find k points that minimize their average pairwise L1 distance. We present a natural approximation algorithm and show that it is a 7/4-approximation for 2D grids. For d-dimensional space, the approximation guarantee is 2-(1/2d), which is tight. We also give a polynomial-time approximation scheme (PTAS) for constant dimension d, and report on experimental results.

cs.DS↗