Chetraj Pandey

Dr. Chetraj Pandey is an Assistant Professor of Computer Science (AI & ML) at Texas Christian University (TCU), specializing in explainable artificial intelligence (XAI), deep learning, and rare event prediction. He earned his Ph.D. in Computer Science from Georgia State University, where his research focused on the development of interpretable deep learning methods for data-intensive applications, including AI systems for high-stakes applications such as space weather forecasting.

Dr. Pandey’s research centers on the integration of explainable AI, continual learning, and space weather forecasting. His doctoral work introduced novel full-disk and active region-based solar flare prediction models, enhancing forecasting capabilities, particularly in near-limb regions that are traditionally difficult to model due to projection effects. He also developed heterogeneous ensemble models that integrate time series and image-based inputs to improve prediction accuracy and granularity. In addition to improving predictive performance, his research introduced techniques for analyzing explanation consistency, investigating multi-granular explanations as proxies for localized predictions, and conducting quantitative evaluations of explanations through physically meaningful features, revealing systematic patterns that offer insight into model behavior. Building on this prior work, Dr. Pandey served as a Scientific Researcher with the NASA-partnered Frontier Development Lab (FDL-X 2024), where he was responsible for the design and implementation of a continual learning framework for forecasting geomagnetic perturbations. This effort contributed to the development of two machine learning pipelines, DAGGER++ and SHEATH, designed to meet operational forecasting requirements, where Dr. Pandey led the continual learning component of DAGGER++ and supported the redevelopment of the SHEATH model to enable integration of real-time solar and geomagnetic data for dynamic forecasting.

At TCU, Dr. Pandey aims to lead research efforts in interpretable and adaptive machine learning, with a focus on continual learning methods that allow AI systems to evolve alongside changing data environments. He looks forward to mentoring students and building collaborative initiatives in trustworthy AI, with applications spanning space weather forecasting and other domains where explainability, adaptability, and reliability are essential.

Education

Ph.D., Computer Science — Georgia State University, 2025

Areas of Focus

Explainable Deep Learning, Pattern Recognition, Continual Learning, Space Weather Forecasting