Khotso Selialia

Khotso Selialia

Auto-generated profile, based on publicly available information

Khotso Selialia

Electrical and Computer Engineering, University of Massachusetts Amherst

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 June 2024
Verified
Conference Paper
Open access
KSKhotso Selialia
Khotso Selialia
YCYasra Chandio
Yasra Chandio
FAFatima M. Anwar
Fatima M. Anwar
Federated learning is emerging as a privacy-preserving model training approach in distributed edge applications. As such, most edge deployments are heterogeneous in nature, i.e., their sensing capabilities and environments vary across deployments. This edge heterogeneity violates the independence and identical distribution (IID) property of local data across clients. It produces biased global models, i.e., models that contribute to unfair decision-making and discrimination against a particular community or a group. Existing bias mitigation techniques only focus on bias generated from label heterogeneity in non-IID data without accounting for domain variations due to feature heterogeneity. Our work proposes a group-fair FL framework that minimizes group-bias while preserving privacy. Our main idea is to leverage average conditional probabilities to compute a cross-domain group importance weights derived from heterogeneous training data to optimize the performance of the worst-performing group using a modified multiplicative weights update method. Additionally, we propose regularization techniques to minimize the difference between the worst and best-performing groups while ensuring through our thresholding mechanism to strike a balance between bias reduction and group performance degradation. Our evaluation of image classification benchmarks assesses the fair decision-making of our framework in real-world settings.

12 December 2025 in ACM Transactions on Sensor Networks
Verified
Journal Article
Open access
YCYasra Chandio
Yasra Chandio
KSKhotso Selialia
Khotso Selialia
JDJoseph DeGol
Joseph DeGol
LGLuis Garcia
Luis Garcia
FAFatima Anwar
Fatima Anwar
Advances in tracking algorithms have enabled applications from steering autonomous vehicles to guiding robots to enhancing augmented reality experiences for users. However, these algorithms are application-specific and do not work across applications with different types of motion; even a tracking algorithm designed for a given application does not work in scenarios deviating from highly standard conditions. For example, a tracking algorithm designed for indoor robot navigation will not work for tracking the same robot in an outdoor environment. To demonstrate this problem, we evaluate the performance of the state-of-the-art tracking methods across various applications and scenarios. To inform our analysis, we first categorize environmental, locomotion-related, and algorithmic challenges faced by tracking algorithms. We quantitatively evaluate the performance using multiple tracking algorithms and representative datasets for a wide range of Internet of Things (IoT) and Extended Reality (XR) applications, including autonomous vehicles, drones, and humans. Our analysis shows that no tracking algorithm works across all applications and even in diverse scenarios within the same application. Ultimately, using the insights generated from our analysis, we discuss multiple approaches to improving the tracking performance using input data characterization, leveraging intermediate information, and output evaluation.