Academic Division: Information Engineering
Research group: Computational and Biological Learning
Broadly speaking, I am interested in high-stakes applications of probabilistic machine learning techniques. The consequences of failure for an ML algorithm that monitors the sensors of an aircraft engine are very different from one that recognizes faces in social media photos or recommends music. These critical applications often place a high premium on principled uncertainty estimates, applicability to limited datasets and interpretability: something probabilistic machine learning techniques like Gaussian processes can offer. Marrying these completely data-driven techniques with physical modeling for more robust predictions and sensible extrapolations is also something that intrigues me.
I am a Marie Sklodowska-Curie Early Stage Researcher in the MAGISTER consortium which seeks to utilize machine learning to understand and predict thermoacoustic oscillations in gas turbines, aircraft engines or rocket engines. My job, as I see it, is to serve as a liaison between the probabilistic machine learning group led by my PhD supervisor Professor Carl Rasmussen and the flow instability and adjoint optimization group led by my advisor Professor Matthew Juniper. We are currently looking at data from small-scale combustors in our lab and large-scale combustor data shared by collaborators from German Aerospace Center (DLR) Lampoldshausen, Rolls-Royce Aircraft Engines and General Electric, to explore how ML techniques can use this data to enable both better designs and safe operation for these machines.
I am also a dilettante computational chemist and am curious about how probabilistic machine learning can improve our ability to predict protein aggregation and protein dynamics. Protein aggregates, of course, play both functional and pathological roles in the human body while protein dynamics is crucial to the functioning of many enzymes. Compared to the static structure prediction problem, however, both aggregation and dynamics are harder to characterize experimentally and lack extensive databases. Can Bayesian techniques shine in this data-limited regime and achieve results comparable to expensive simulations which consume many thousand of supercomputer core-hours?
- Demonstrator for 3F8 Inference (Lent 2019)
- Demonstrator for 1A Dimensional Analysis Fluids Lab (Michaelmas 2020, Michaelmas 2021)
- Demonstrator for 1B Boundary Layer Lab (Michaelmas 2020)
- Co-supervising Rami Cassia's Master Thesis on "Flow and Shape Inference in Magnetic Resonance Velocimetry using Physics-Informed Neural Networks" (2021)
I did my bachelor in Mechanical Engineering from the Indian Institute of Technology, Kharagpur and my masters in computational science from RWTH Aachen University, Germany. As a bachelor student, I worked on the computational modeling of compartment fires and microcombustors. In my masters thesis work, on the other hand, I analyzed data from molecular dynamics simulations and focused on the automatic generation of hidden Markov models to help computational scientists effortlessly derive a simple, concise "states and rates" picture from the massive amount of data they generate. My experiences shaped me into a person with great passion for both mechanical engineering and data science and I believe that the topic of my PhD represents a perfect fusion of these interests.