QUEST Lab — IISc

Research

From ocean currents to elephant migrations, the QUEST Lab keeps returning to one question: how do you build machine learning that respects the physics, uncertainty, and constraints of the real world? Seven threads of work — and a few new directions — answer that across very different domains.

Physics-Informed ML & Neural Operators

Neural operators, PINNs, and diffusion models that bake in physics — from wind-turbine wakes to SAR object detection.

Ocean, Weather & Climate Prediction

Probabilistic ocean forecasting, climate-model bias correction, and coastal eddy/current detection in the Indian Ocean and beyond.

Robotics and Physical Intelligence

Optimal path planning for marine and aerial vehicles under uncertainty, now growing into Physical AI — the lab's primary focus: World Models, VLAs, VLMs, LLMs, and Flow Models.

Indian Monsoon & Weather Extremes

Forecasting Indian Summer Monsoon onset, active-break spells, and city-scale extreme rainfall.

Data Assimilation, Bayesian Methods & Uncertainty Quantification

The methodological thread underneath the lab's ocean, monsoon, and epidemiological modeling: filtering theory and uncertainty quantification.

AI for Social Good: Ecology & Conservation

Agent-based models and habitat prediction for human-wildlife conflict, plus sustainable-fisheries modeling.

AI for Education

LLM-based question generation and speaking-assessment scoring, aligned to Bloom's taxonomy and CEFR standards.

Emerging Directions

What's next for the lab — the new primary research focus, plus other early-stage directions we're actively building toward, ahead of the papers that will follow.