Interdisciplinary Applications of Physics and Machine Learning

My research interests span Physics, Machine Learning (ML), and Quantum Cryptography and focus on three areas:

1. Raman spectroscopy combined with ML holds great promise for many applications as a rapid, label-free identification method and works well when classifying spectra of chemicals encountered during training. However, in real-world conditions, such as clinical applications, there are always species whose spectra have not yet been taken.

Simply put, if the ML system is trained to identify either “cats" or “dogs" but sees a new sample of “fish" during testing, the false-positive rate becomes uncontrollable and the NN will misclassify “fish" as either “cat" or “dog". This limits the usefulness of these techniques, especially in public safety applications. To overcome these barriers, I combined the Objectosphere loss function with ResNet architecture to show that this approach allows for separation of the unknowns while remaining highly accurate on the known species and performs better than the current gold standards in ML techniques.

This opens the door to using Raman spectroscopy, combined with our novel machine learning algorithm, in a variety of practical applications.

2. Up to now, Dark Matter (DM) manifests itself only by astronomical observations. To reveal its nature, it is necessary to find it through experiments here on Earth. DM particles may interact with the visible particles, and new physics can be searched from a small mismatch between the Standard Model theoretical predictions and experimental measurements. In my work, the calculations needed to accomplish this were successfully performed.

3. Quantum Computers are becoming a reality. We developed a framework for the analysis of security and performance of post-quantum signature algorithms. Additionally, we developed applications of a novel approach called PT symmetry for the purposes of Quantum Cryptography and performed proof-of-concept experiments using IBM Quantum Experience.