Email: aakashbhatia19@gmail.com
Website: https://www.github.com/aakashb1
Link to CV: https://drive.google.com/drive/folders/1l18yBnoB4GEvoUZY2nZxmmhIe3TAUtRt?usp=sharing
Level: PhD, MS, Internship, Research Visit
Keywords: Python, Data Science, Machine Learning
Short bio / research interests / publications: I was born in a family, where I was given liberty to pursue my interests which gave me the space to think freely and ask questions. From diving into textbooks to getting our hands dirty in the laboratory, I was always fascinated by the parallels we can draw in the two worlds. I studied hard and secured my B.Tech from Malviya National Institute of Technology Jaipur where I had the opportunity to study mathematics, thermodynamics, simulation etc.
In 2017, I moved to Pittsburgh to pursue my Masters of Sciences from Carnegie Mellon University which was a mix of course work and thesis. I chose Chemical Engineering because I was really fascinated by the research on mathematics performed by the faculty there. From attending lectures on optimizing large scale systems by Prof. Larry Biegler to implementing genetic algorithms to solve an NP-hard problem, the coursework really helped me build a foundation. Later, I also got the opportunity to work on my master’s thesis under the guidance of Prof. Nikolaos Sahinidis on an industrial text analytics problem for reliability. I also got an opportunity to take courses on deep learning and natural language processing which further helped me enhance my working knowledge on my thesis. The course also helped me get hands on experience on both text and utterances as input.
Adding to the academic background, I was involved in various hackathons that involved problems from the medical domain. One of which was to identify sleeping patterns in a healthy human’s sleeping cycle for which we in a team of 5 built a binary classification model utilizing a sliding window approach to feature engineer the data. The second was a natural language problem to identify the cancer site location based on a person’s diagnostic history, the event was in collaboration with a hospital in Pittsburgh.
At present, I hold a great deal of interests in machine learning. For example trying to understand how state of the art machine learning techniques can help gain better insights on underlying operational data. Although neural networks are tagged as universal approximators, one can still see varying performances in the domain of images and natural language. While challenging, I feel I would be able to contribute to the lab’s ongoing research with my interdisciplinary research and more than that learn the missing connections through the use of algorithmic techniques that form the core of many machine learning models out there.