ODDESSY

ODDESSY Lab

Optimization, Data and Decision Systems

We develop rigorous mathematical foundations and scalable algorithms at the intersection of optimization, stochastic systems, game theory, and data-driven control — with applications to networked infrastructure, epidemic dynamics, and multi-agent systems.

Department of Electrical Engineering  ·  IIT Kharagpur  ·  PI: Ashish R. Hota

Our goal is to understand and shape decision-making in complex, uncertain, and interconnected systems — bridging the gap between mathematical theory and real-world engineering challenges through principled data-driven methods.

— Mission of the ODDESSY Lab

Research Focus Areas

Stochastic Optimization & Control

We design optimization and control algorithms that are robust to uncertainty without requiring full distributional knowledge. Our work on distributionally robust and data-driven methods provides rigorous performance guarantees using only observed data, with applications in power systems, robotics, and multi-robot navigation.

Distributionally Robust Opt. Wasserstein Ambiguity MPC Risk Measures

Game Theory & Network Security

We analyse strategic interactions among rational and boundedly-rational agents on networks. Our research quantifies how decentralised decision-making affects security investments, resource sharing, and systemic resilience in interdependent infrastructure.

Nash Equilibrium Attack Graphs Prospect Theory Network Resilience

Epidemics on Networks: Modeling & Control

We combine network science, dynamical systems, and game theory to model the spread of infectious diseases and opinions. We study how individual protective behaviour evolves under uncertainty and design optimal non-pharmaceutical interventions using closed-loop feedback.

SIS / SIR Models Population Games Network Estimation Behavioral Biases

Behavioral Economics & Mechanism Design

We investigate how cognitive biases and prospect-theoretic preferences shape human interaction with shared engineered systems. We design dynamic incentive mechanisms — taxes, subsidies, and information signals — to align individual and societal objectives.

Mechanism Design Incentive Engineering Human-in-the-Loop Smart Grid

Learning in Multi-Agent Systems

We study how agents learn and adapt strategies in non-stationary, adversarial, and cooperative environments. Our work develops convergent online learning algorithms and population game dynamics with provable stability and regret guarantees.

Reinforcement Learning Online Learning Evolutionary Dynamics Multi-Agent RL

Power Systems & Smart Infrastructure

We apply distributionally robust optimisation and game-theoretic tools to energy systems, enabling reliable and economically efficient dispatch, demand response, and risk-aware control under the uncertainty brought by renewable generation.

Economic Dispatch Demand Response Renewable Integration Robust Control

Principal Investigator

AH

Ashish R. Hota

Associate Professor & Lab Director

Department of Electrical Engineering, IIT Kharagpur  ·  Ph.D. Purdue University (2017)  ·  Postdoc ETH Zürich (2018)  ·  Young Associate, Indian National Academy of Engineering (2023)

→ Full profile    [email protected]