We are a data sciences research and teaching centre. Our primary goal is to support effective decision making related to the complex challenges faced by society.
The growing data, compute and algorithmic advancements are rapidly and profoundly transforming the world. And this is early days in the data science, machine learning, artificial intelligence driven revolution! At SCDLDS we conduct applied and methodological research in data and learning sciences at the highest level to aid in informed decision making to address the complex challenges faced by the society.
Initially our focus is on four research areas: Data driven financial research, weather and climate modelling, reinforcement learning and epidemiological modelling.
The centre will leverage Ashoka’s interdisciplinary strengths in computer sciences, mathematics, economics, physical and social sciences to foster interdisciplinary collaborations that creatively interpret and analyze data to solve societal problems. It will help train young minds through exciting course work and projects and expose them to a deep and varied mathematical toolkit to uncover hidden patterns in large and complex datasets.
|
We look at many of the important challenges in financial mathematics through the lens of latest developments in machine learning. Some of the problems of interest include data driven portfolio risk measurement including market, credit and operational risk for large banks. Portfolio optimization while managing the return-risk trade off. Pricing and hedging high dimensional options. Measuring systemic risk in the banking system, and so on.
We are developing machine learning based models that combine historical data with output from physics based numerical weather prediction models to improve monsoon weather prediction in India. We are also using ideas from extreme value theory to develop algorithmic methodologies to predict weather related extreme events including high precipitation, heat waves, cyclones, flooding, etc.
Reinforcement learning finds wide applications in diverse areas including healthcare, autonomous vehicles, robotics, financial trading and portfolio risk measurement, traffic modelling. In the center we focus on methodological issues associated with the underlying reinforcement learning models to arrive at accurate and computationally efficient algorithms for massive RL models, exploiting latest developments in machine learning.
Infectious diseases such as Covid have been a subject of great deal of analysis through mathematical modelling, simulation and machine learning models. In the centre we will collect relevant data and develop cutting edge agent based simulation models models to help study disease spread and predict its evolution trajectory.