** I am currently open to opportunities in the roles of an Applied Scientist or Machine Learning Researcher. I welcome any interested parties to connect with me on LinkedIn or via my email at email@example.com. **
Possessing a rich background in Machine Learning research, I bring to the table comprehensive expertise in Convex, Non-Convex, and Constrained Optimization, Causal Inference, and Deep Learning. My technical acumen shines in my ability to design and implement Convex and Non-Convex solvers which streamline the training process of machine learning models. In this capacity, I have accumulated vast experience with open-source platforms like Pytorch and Tensorflow, allowing me to modify and even extend their functionalities.
In the healthcare industry, my work has been pivotal in utilizing constrained optimization solvers for disease phenotyping. This work has paved the way for an innovative interpretable treatment recommendation model.
I’ve also made significant contributions in the development of a causal reinforcement learning algorithm. This unique algorithm leverages causal inference principles to forecast more accurate potential treatment outcomes.
My involvement in Deep Learning has resulted in the creation of a model that outperforms the current state-of-the-art models in predicting regulatory elements.
Recently, I’ve further refined my optimization skills by creating a constrained optimization solver. This solver can identify pivotal features corresponding to multidimensional feature and outcome spaces. Recognized by experts in the field, this represents a high-dimensional extension of L1 lasso regression, thereby enhancing the model’s performance and explainability significantly.
At present, my focus is on designing cutting-edge deep learning models to solve intricate Natural Language Processing problems. A notable accomplishment is the unique framework I’ve developed for news framing prediction. This framework adopts a graph-based approach that converts sentences into graph nodes, utilizes transformer models for embeddings, and extracts relationships from the ASER Knowledge Graph. A trained Graph Neural Network is then deployed to fine-tune transformers and accurately predict the framing of news articles.
In addition to my research work, I serve as a conference reviewer for highly respected forums such as NeurIPS and ICML, contributing to the shared body of knowledge in this rapidly progressing field. In my upcoming role, I aim to utilize my specialized expertise to continue propelling the boundaries of research forward.
- Invited to serve as a reviewer of NeurIPS. March, 2023.
- Invited to serve as a reviewer of ICML. Jan, 2023.
- Postdoctoral Research Associate at Purdue University , Jan, 2023.
- Ph.D Graduated, Dec, 2022.
- Paper accepted in PLOS Digital Health, Sep ‘22.
- Invited to serve as a reviewer of NeurIPS. March, 2022.
- Invited to serve as a reviewer of ICML. Feb, 2022.
- Invited to serve as a reviewer of ICPP. May, 2021.
- Presented our work “Identifying and Analyzing Sepsis States: A Retrospective Study on Patients with Sepsis in ICUs” at Regenstrief Center for Healthcare Engineering (RCHE), (flyer). Feb 24, 2021.
- Featured article about our Distributed GPU-Accelerated Optimizer for Multiclass Classification Problems (link). Feb 3, 2021.
- Google Fellowship Nominated by Purdue University. Sept, 2020
- Invited to serve as a reviewer of ICDM. 2020.
- Book Chapter “Parallel Optimization Techniques for Machine Learning”. Springer. 2020
- Received Regenstrief Center for Healthcare Engineering (RCHE) Student Scholarship, Purdue University. 2020 (Tweet)
- Our paper “Newton-ADMM: A Distributed GPU-Accelerated Optimizer for Multiclass Classification Problems” has been accepted in the Proceedings of the ACM/IEEE Supercomputing Conference (SC20) - 18% acceptance rate.
- Received Regenstrief Center for Healthcare Engineering (RCHE) Travel Award, Purdue University. 2020
- Received SIAM International Conference on Data Mining Travel Award (Cancelled due to COVID-19 outbreak), Cincinnati, Ohio, USA. 2020