QUEST Lab — IISc
Publications
A decade and a half of turning stochastic optimal control, uncertainty quantification, and modern deep learning into results that ship — from a US patent on autonomous vehicle routing to award-winning conservation research. Browse by research theme below, newest work first, or read the story behind each thread on the Research page.
Full citation metrics (citations, h-index, i10-index) on Google Scholar ↗
Physics-Informed ML & Neural Operators
Multi-turbine turbulent wake simulation using physics-informed neural networks
WAKE-NET project
DNOD: Deformable Neural Operators for Object Detection in SAR Images
★ J2C Certification — Top 3% of Accepted Papers
A physics-informed neural network for turbulent wake simulations behind wind turbines
WAKE-NET project
Conditional Diffusion Model with Nonlinear Data Transformation for Time Series Forecasting
★ CORE2023 Ranking: A*. Acceptance rate = 26.9%
CNDiff
Discrete Residual Loss Functions for Training Physics-Informed Neural Networks
On the Training Efficiency of Shallow Architectures for Physics Informed Neural Networks
Ocean, Weather & Climate Prediction
High-resolution simulation of the Bay of Bengal rain plume
High frequency radar observations of east India coastal current eddies
Global climate model bias correction using deep learning
Cause and impact of Andaman Sea's salinity variability: A modeling study
A two-phase Neural Model for CMIP6 bias correction
AI-based correction of CMIP6 ocean projections
Variations of Eddy Characteristics Along the East Coast of India
4D-Var Data Assimilation of Sea Surface Temperature in a Regional Model of the Andaman Sea
Inter-Annual Variability of Monsoon and Post-Monsoon Salinity in Andaman Sea
Super Resolution of Sea Surface Temperature Satellite Images
Deep Learning for Detecting Gulf Stream and Eddies from Satellite Images
Bayesian Estimation and Data Assimilation for Probabilistic Regional Forecasts in the northern Indian Ocean
Optimal Planning and Sampling Predictions for Autonomous and Lagrangian Platforms and Sensors in the Northern Arabian Sea
Stochastic Ocean Predictions with Dynamically-Orthogonal Primitive Equations
Energy Optimal Path Planning by Stochastic Dynamically Orthogonal Level-Set Optimization
PhD thesis work
Probabilistic Predictions of Regional Ocean Dynamics
Probabilistic Regional Ocean Predictions
Robotics and Physical Intelligence
Physics-Guided Encoder-Decoder Transformers for Path Planning of Autonomous Marine Vehicles
Intelligent Onboard Routing in Stochastic Dynamic Environments using Transformers
★ CORE2023 A* Ranking. Acceptance Rate = 23%
GPU-Accelerated Multi-Objective Optimal Planning in Stochastic Dynamic Environments
Physics-Driven Machine Learning for Time-Optimal Path Planning in Stochastic Dynamic Flows
Risk Optimal Path Planning in Stochastic Dynamic Environments
Stochastic Time Optimal Path Planning in Uncertain, Strong, and Dynamic Flows
Time-Optimal Multi-Waypoint Mission Planning in Dynamic Environments
Clustering of Massive Ensemble of Vehicle Trajectories in Strong, Dynamic and Uncertain Ocean Flows
Probabilistic Risk-Optimal Path Planning and Sampling in Uncertain Environments
A Future for Intelligent Autonomous Ocean Observing Systems
Multiple-Pursuit/One-Evader Pursuit Evasion Game in Dynamical Flow Fields
Pursuit-Evasion Games in Dynamic Flow Fields via Reachability Set Analysis
Data-driven Learning and Modeling of AUV Operational Characteristics for Optimal Path Planning
Energy Optimal Path Planning: Integrating Coastal Ocean Modelling with Optimal Control
A stochastic optimization method for energy-based path planning
Indian Monsoon & Weather Extremes
A machine learning model for active-break spell forecasting of Indian Summer Monsoon from Outgoing Longwave Radiation data
Comparison of Convolutional Autoencoder Architectures for Representation Learning of MODIS Aqua and Terra Observations
Data Assimilation, Bayesian Methods & Uncertainty Quantification
Spatio-temporal predictive modeling framework for infectious disease spread
A new ensemble-based data assimilation algorithm to improve track prediction of tropical cyclones
On the Effect of Non-Raining Parameters in Retrieval of Surface Rain Rate Using TRMM PR and TMI Measurements
AI for Education
EvalYaks: Instruction tuning datasets and LoRA fine-tuned models for automated scoring of CEFR B2 speaking assessment transcripts
★ CORE2023: A Ranking, Accepted as Main Paper (Oral). Acceptance rate = 23%
How Good are Modern LLMs in Generating Relevant and High-Quality Questions at Different Bloom's Skill Levels for Indian High School Social Science Curriculum?
Preprints
Not yet peer-reviewed.