Research | Chetraj Pandey

My research connects interpretable and adaptive machine learning with the challenges of scientific forecasting.

Temporal Tracking of a X-class flare on the Eastern-limb of the Sun with corresponding explanation maps.

Interpretability

Explainable artificial intelligence

Understanding why a model predicts an event—and whether its reasoning is meaningful.

Schematic Diagram of DAGGER in CL paradigm.

Continual learning

Learning in changing environments

Building models that can adapt as new observations arrive and data distributions shift.

X1.4-class solar flare observed on September 22, 2011, by NASA's Solar Dynamics Observatory (SDO) using the Atmospheric Imaging Assembly (AIA).

Scientific machine learning

AI for space weather

Forecasting solar flares and geomagnetic disturbances from complex scientific observations.

Connecting methods and applications

These directions are closely connected: a useful forecast needs reliable predictions, meaningful explanations, and the ability to respond to changing observations.