The Center for Research toward Advancing Financial Technologies presents the CRAFT Student Network. The following students have worked alongside principal investigators on NSF-funded I/UCRC CRAFT Fintech Research Projects.
Nan Cui
CRAFT Research Project: Federated Learning for Fairness-aware and Privacy-Preserving Financial Risk Assessment
Educational Background
Ph.D. in Computer Science, Stevens Institute of Technology (SIT), Hoboken, NJ (08/2020 - Present)
M.S. in Computer Science, Stevens Institute of Technology (SIT), Hoboken, NJ (01/2019 - 05/2020)
M.S. in Electrical Engineering, University of Massachusetts Lowell (UMass Lowell), Lowell, MA (08/2015 - 12/2017)
B.E. in Electrical Engineering and Automation, Northeast Electric Power University (NEEPU), Jilin, China (08/2011 - 07/2015)
Bio
Nan Cui is a Ph.D. student in Computer Science at Stevens Institute of Technology, working under Professor Yue Ning. She has a strong background in Computer Science and Electrical Engineering. She has worked on several projects, including the development of a federated learning framework tailored to support fairness-aware Graph Convolutional Networks (GCNs) and a fair and efficient active learning framework for designing label-efficient algorithms.
Research Interests
Graph Deep Learning
Federated Learning
Fairness in Machine Learning
Publications
Metric-Fair Active Learning. Jie Shen, Nan Cui, and Jing Wang. In Proceedings of the 39th International Conference on Machine Learning, 2022.
Applying Gradient Descent in Convolutional Neural Networks. Nan Cui. In Proceedings of the 2nd International Conference on Machine Vision and Information Technology, 2018.
History of Research Projects
Fair Graph Neural Networks in Federated Learning
Individual Fairness in Active Learning
Contact
Zhiyang Deng
CRAFT Research Project: Fast Quantum Methods for Financial Risk Management
Educational Background
Ph.D. Student in Financial Engineering, Stevens Institute of Technology
M.S. in Financial Mathematics, University of Southern California (USC)
B.S. in Mathematics and Applied Mathematics, Nanchang University
Bio
Zhiyang Deng is a PhD candidate specializing in Financial Engineering at Stevens Institute of Technology. With a strong foundation in mathematics, Deng demonstrates proficiency in complex financial models and quantitative analysis. His research interests focus on Stochastic Control, Mean Field Games, Optimization in Finance.
Deng has received Provost's Doctoral Fellowships from Stevens due to the recognition from the academic community. Outside his academic pursuits, he's an active participant in industry, and he currently serves as a fixed income Desk Quant intern at RBC Capital Markets.
For further updates or collaboration opportunities, you can connect with him on LinkedIn.
Research Interests
Stochastic Control
(Numerical) Mean Field Games
Risk Measure
Reinforcement Learning and Optimal Control
Optimization in Finance
History of Research Projects
Risk Averse Mean Field Games and Delta Dirac Family Methods
Deep learning formulation via optimal control perspective
Fast Quantum Methods for Financial Risk Management
Contact
Conor Flynn
CRAFT Research Project: DeFi Data Engine (May 2022 - May 2023)
Educational Background
Incoming Ph.D. Student in Computer Systems Engineering at Rensselaer Polytechnic Institute
Bachelor of Computer Science with a minor in Math at Rensselaer Polytechnic Institute, Class of 2023
Bio
Conor Flynn is an incoming Ph.D. student at Rensselaer Polytechnic Institute studying computer systems engineering. Having studied programming since middle school, Conor has since participated in various clubs, coding competitions, and leadership positions revolving around his love for computer science.
Throughout his professional journey, Conor has since worked as an intern as well as an entrepreneur. In 2019, he worked his first professional position as an Information Security Intern at PDX Inc., helping to secure their networking systems. In 2022, Conor was a Computer Science Intern at Convergint Technologies LLC, working on computer vision modeling for their camera systems to help improve their existing infrastructure. His entrepreneurial ventures started in 2020 with the founding of his first company Quantify Enterprises LLC. Throughout his duration working as its sole proprietor, Conor was contracted by various financial professionals and institutions to help improve their existing quantitative trading systems and strategies. Following his success at Quantify, he partook in the founding of ScaleTrade LLC with two colleagues in 2022, aiming to help traders improve their returns through the generation of unique and insightful technical data.
Conor also has done undergraduate research under the supervision of Professors Kristin Bennett and Professor Oshani Seneviratne starting in May of 2022 through May of 2023. During this time, he worked on a programming language and data agnostic, low latency data engine referred to as the DeFi Data Engine. This engine aims to improve the data collection and distribution for the lab for future research.
Research Interests
Low Latency Programming
Computationally Agnostic System Designs
Quantitative Finance
Deep Learning
Applied Mathematics
Publications
Adams, Kacy, Fernando Spadea, Conor Flynn, and Oshani Seneviratne. "Assessing Scientific Contributions in Data Sharing Spaces." arXiv preprint arXiv:2303.10476
(2023).
Contact
Michael Giannattasio
CRAFT Research Project: Risky Business? Deep Dives into DeFi
Educational Background
Rising Senior at Rensselaer Polytechnic Institute with a major in Mathematics
Bio
My name is Michael Giannattasio and I am a rising senior at Rensselaer Polytechnic Institute. I have a major in Mathematics specializing in Operations Research with a minor in Economics of Quantitative Modeling. I have been working on the CRAFT project Risky Business? Deep Dives into DeFi from the Summer of 2022 to the present. I have been focused on user-level representation of transactional data and discovering novel user-clustering methods, specifically across multiple datasets, to apply to multiple DeFi ecosystems' users at once. In order to display and facilitate the visualizations of these user clusters along with survival analysis results, I have been developing the DeFi Survival Analysis Toolkit with authored documentation located here.
Research Interests
Novel Clustering Methods
Risk Analysis
Data Science
Machine Learning
Artificial Intelligence
Publications
DeFi Survival Analysis: Insights Into the Emerging Decentralized Financial Ecosystem (Accepted by ACM-DLT); Characterizing Common Quarterly Behaviors in DeFi (Accepted by MARBLE 2023)
Contact
Inwon Kang
CRAFT Research Project: Risk Mitigation in Cross-Platform Decentralized Finance
Educational Background
Ph.D. Candidate, Computer Science, Rensselaer Polytechnic Institute (Current)
M.S., Computer Science, Rensselaer Polytechnic Institute (May 2022)
B.S., Computer Science, Rensselaer Polytechnic Institute (May 2020)
Bio
I am a current Ph.D. student at Rensselaer Polytechnic Institute, working under Professor Oshani Seneviratne. I work mostly with Python for data/analysis oriented projects. I am also familiar with javascript-based frameworks. During my undergraduate and master's program, I focused on explainable machine learning using shallow models. My master's project was on collecting survey data from participants and building explainable ML models to try to explain the results. During this time, I also dabbled in NLP and built an enhanced pipeline of existing state-of-art model for Preference Ellicitation task using GNNs and coreference parsers. My Ph.D. focus is on blockchain systems and integration of machine learning with blockchains.⠀⠀⠀
Research Interests
Blockchain Interoperability
System Design
Deep Learning
Natural Language Processing
Publications
Dependency and Coreference-boosted Multi-Sentence Preference Model. Farhad Mohsin, Inwon Kang, Yuxuan Chen, Jingbo Shang and Lirong Xia. (DLG-AAAI’23)
Learning to explain voting rules. Inwon Kang, Qishen Han, Lirong Xia (Extended Abstract, AAMAS 2023)
Analyzing and Predicting Success in Music. Inwon Kang, Michael Manduluk, Boleslaw Szymanski. (Scientific Reports)
Blockchain Interoperability Landscape. Inwon Kang, Aparna Gupta, Oshani Seneviratne. (IEEE BigData Distributed Storage Workshop 2022)
Landslide Likelihood Prediction using Machine Learning Algorithms. Vasundhara Acharya, Anindita, Inwon Kang, Thilanka Munasinghe, Binita KC (IEEE BigData 2022)
Crowdsourcing Perceptions of Gerrymandering. Benjamin Kelly, Inwon Kang, Lirong Xia. (HCOMP 2022)
Learning Individual and Collective Priorities over Moral Dilemmas with the Life Jacket Dataset. Farhad Mohsin, Inwon Kang, Pin-Yu Chen, Francesca Rossi and Lirong Xia. (MPREF-22)
Making group decisions from natural language-based preferences. Farhad Mohsin, Lei Luo, Wufei Ma, Inwon Kang, Zhibing Zhao, Ao Liu, Rohit Vaish and Lirong Xia. (COMSOC-21)
History of research projects
Contact
Md. Saikat Islam Khan
CRAFT Research Project: Efficient, Private, and Explainable Federated Learning for Financial Crime Detection
Educational Background
Ph.D. Candidate, Computer Science, Rensselaer Polytechnic Institute (August 2023 – May 2028)
B.Sc. in Computer Science and Engineering (February 2015 – January 2020)
Bio
I am currently a Ph.D. student at Rensselaer Polytechnic Institute, working under Professor Oshani Seneviratne. My research activities span a broad spectrum of applications in machine learning, computer vision, and data science. During my undergraduate studies, I developed various classification models to classify medical images. My Ph.D. focus is on federated large language models and explainable federated learning. Through my involvement with CRAFT, I introduced Federated Learning for Relational Data (Fed-RD), a novel privacy-preserving federated learning framework specifically developed for financial transaction datasets that are partitioned both vertically and horizontally across parties. Fed-RD strategically employs differential privacy and secure multiparty computation to ensure the privacy of training data. I now aim to improve this project by enhancing the explainability of the black-box model.⠀⠀⠀⠀⠀⠀⠀⠀⠀
Research Interests
Distributed Systems
Federated Large Language Model
Explainable Federated Learning
Health Informatics
Publications
Fed-RD: Privacy-Preserving Federated Learning for Financial Crime Detection MSI Khan, A Gupta, O Seneviratne, S Patterson - arXiv preprint arXiv:2408.01609, 2024
A differentially private blockchain-based approach for vertical federated learning L Tran, S Chari, MSI Khan, A Zachariah, S Patterson… - arXiv preprint arXiv:2407.07054, 2024
Accurate brain tumor detection using deep convolutional neural network MSI Khan, A Rahman, T Debnath, MR Karim, MK Nasir… - Computational and Structural Biotechnology Journal, 2022
Water quality prediction and classification based on principal component regression and gradient boosting classifier approach MSI Khan, N Islam, J Uddin, S Islam, MK Nasir - Journal of King Saud University-Computer and …, 2022
MultiNet: A deep neural network approach for detecting breast cancer through multi-scale feature fusion SI Khan, A Shahrior, R Karim, M Hasan, A Rahman - Journal of King Saud University-Computer and …, 2022
Using fused Contourlet transform and neural features to spot COVID19 infections in CT scan images A Nur, MSI Khan, MK Nasir - Intelligent Systems with Applications, 2023
Proposing a hybrid technique of feature fusion and convolutional neural network for melanoma skin cancer detection MM Rahman, MK Nasir, M Nur-A-Alam, MSI Khan - Journal of Pathology Informatics, 2023
BlockSD‐5GNet: Enhancing security of 5G network through blockchain‐SDN with ML‐based bandwidth prediction A Rahman, MSI Khan, A Montieri, MJ Islam, MR Karim… - Transactions on Emerging Telecommunications …, 2024
Machine learning and deep learning-based approach in smart healthcare: Recent advances, applications, challenges and opportunities A Rahman, T Debnath, D Kundu, MSI Khan, AA Aishi… - AIMS Public Health, 2024
Federated learning-based AI approaches in smart healthcare: concepts, taxonomies, challenges and open issues A Rahman, MS Hossain, G Muhammad, D Kundu… - Cluster computing, 2023
Biorthogonal wavelet based entropy feature extraction for identification of maize leaf diseases B Mazumder, MSI Khan, KMM Uddin - Journal of Agriculture and Food Research, 2023
Contact
Tasha Kholsa
CRAFT Research Project: Comprehensive Financial Disclosure Lexicon
Educational Background
B.S. in Quantitative Finance, Stevens Institute of Technology, Class of 2025
Bio
Tasha Khosla is a rising third year quantitative finance student pursuing a concentration in quantitative methods. A passion for mathematics, programming, and finance is what led her to pursue this degree at Stevens Institute of Technology. Outside of academics, she is involved in various clubs and organizations on campus. In her free time, Tasha enjoys reading, writing, and exploring new programming languages and techniques.
An interest in financial technology (fintech) prompted her to work on a CRAFT research project this summer, studying the application of natural language processing (NLP) in finance and accounting research through financial literature reviews. ⠀⠀⠀⠀⠀⠀⠀⠀⠀
Research Interests
Artificial Intelligence
Green Finance,
Machine Learning,
Natural Language Processing
Contact
Dominick Varano
CRAFT Research Project: Federated Learning for Fairness-Aware and Privacy-Preserving Financial Risk Assessment
Educational Background
B.S. in Computer Science at Stevens Institute of Technology
Bio
Dominick Varano is a fourth-year undergraduate student in Computer Science at Stevens Institute of Technology. He has a strong passion for Machine Learning and Artificial Intelligence. His research has focused on studying and developing individual and group fairness algorithms within federated learning frameworks.
Research Interests
Machine Learning
Artificial Intelligence
Natural Language Processing
Contact
Bolun “Namir” Xia
CRAFT Research Project: Predictive Learning from Long Financial Documents
Educational Background
Ph.D. Candidate, Computer Science, Rensselaer Polytechnic Institute (August 2022 - )
M.S., Computer Science, Rensselaer Polytechnic Institute (August 2020 - May 2022)
B.S., Finance and Computer Science, New York University, Stern School of Business (Fall 2016 - May 2020)
Bio
Bolun "Namir" Xia is a Ph.D. candidate in Computer Science at Rensselaer Polytechnic Institute, with an undergraduate background in Finance and Computer Science. His research focuses on Natural Language Processing (NLP) and its real-world applications in Finance. Being both versed in Finance and Computer Science, he aims to bridge the gap between both, contributing to the growing field of Financial Technologies in the NLP sphere. His work mainly focuses on developing better representations of textual data for predictive analytics, with cutting-edge NLP methods such as Transformer Language Models and Graph Deep Learning. Through his involvement with CRAFT, he has been able to produce favorable results in long document regression tasks, which predict numerical target variables using long documents, such as using 10-K reports to predict firm performance indicators and analyzing earnings calls to forecast its immediate impact in temporal stock price dynamics. He now aims to improve the craft by designing novel graph-based deep learning methods specialized for prediction tasks based on financial text. In his free time, he likes to practice traditional archery and learn different languages.
Research Interests
Natural Language Processing
Predictive Analytics
Long Document Regression
Graph Deep Learning
Machine Learning
Transformer Language Models
Publications
Bolun "Namir" Xia, Vipula D. Rawte, Mohammed J. Zaki, and Aparna Gupta. FETILDA: an effective framework for fin-tuned embeddings for long financial text documents. arXiv Computing Research Repository, Jun 2022.
History of research projects
Predictive Learning from Long Financial Documents