Curriculum Vitae
Alexander Vidal
PhD Candidate and Senior Data Scientist
alexanderrobertvidal@gmail.com
Education
- Ph.D. in Applied Mathematics and Statistics, 2019-May 2024
Department of Applied Mathematics and Statistics, Colorado School of Mines M.Sc. in Applied Mathematics and Statistics, Minor: Electrical Engineering, 2018-2019
Department of Applied Mathematics and Statistics, Colorado School of Mines- B.Sc. in Engineering, Minor: Economics, 2006-2010
University of Colorado - Boulder
Experience
- Senior Data Scientist, On The Barrelhead / NerdWallet, October 2021 - Present
- Working group: Credit Cards and Lending
- Used data science to solve problems and help make business decisions:
- Lead implementation of recommender algorithms for all credit card pages.
- Predictive modeling to help determine correct product or flow.
- Portfolio optimization to help determine optimal ratio of products to sell to users.
- Generative modeling to bolster class-imbalanced datasets with additional synthetic data.
- Chief Data Scientist, Rigorous Machine Learning Solutions, LLC, October 2022 - Present
- National Science Foundation (NSF) Intern, USGS, June 2021-August 2021
- Worked with the USGS hyperspectral team to collect hyperspectral data for more accurate predictive analysis.
- Graduate Research Assistant, CASERM, 2019 - 2021
- Collected two different types of mineralogical data and applied image recognition techniques in order to reconcile the two datasets.
- A preprocessing step was applied that uses convolutional neural networks to “mask” the data that is not useful.
- A stochastic autoencoder (SAE) is used to ‘mix’ the data used from different sources a latent space.
- Neural network is used to allow for prediction of one dataset given the other.
- Data Science Intern, Lumen Technologies (formerly CenturyLink), June 2019 - August 2019
- Working Group: Finance
- Classified pdf documents using deep learning and natural language processing (NLP).
- Teaching Assistant, Colorado School of Mines, 2018-2021
- Classes: MATH534/535: Mathematical Statistics, MATH537: Multivariate Analysis, MATH536: Advanced Statistical Modeling, MATH225: Differential Equations.
Computer Skills
- Programming languages: Python, R, Matlab, Bash/Shell script, LaTeX, PostgreSQL, MSSQL
- Packages: Numpy, Scipy, Pandas, scikit-learn, Pytorch, Keras, Tensorflow, CVX, CVXPy
- Operating Systems: MacOS, Linux, Windows
- Other: Git
Software Publications and Contributions
- Python: JKO-Flow (Pytorch implementation for the JKO-Flow paper)
Peer-Reviewed Publications
Interpretation of Hyperspectral Shortwave Infrared Core Scanning Data Using SEM-Based Automated Mineralogy: A Machine Learning Approach
Rotem, Amit, Vidal, Alexander, Pfaff, Katharina, Tenorio, Luis, Chung, Matthias, Tharalson, Erik, Monecke, Thomas: Journal of Geosciences, DOI: https://www.mdpi.com/2076-3263/13/7/192 , 2023.
Taming hyperparameter tuning in continuous normalizing flows using the JKO scheme
Vidal, Alexander, Wu Fung, Samy, Tenorio, Luis, Osher, Stanley, Nurbekyan, Levon: "Taming hyperparameter tuning in continuous normalizing flows using the JKO scheme", Scientific Reports, DOI: https://www.nature.com/articles/s41598-023-31521-y, 2023.
Manuscripts in Preparation and Preprints
Kernel Expansions for Mean-Field Control
Vidal, Alexander , Wu Fung, Samy, Tenorio, Luis, Osher, Stanley, Nurbekyan, Levon
Conference Contributions & Talks
Kernel Expansions for Mean Field Control
Talk at Kernel Club Reading Group, Colorado School of Mines
Taming hyperparameter tuning in continuous normalizing flows using the JKO scheme (Invited)
Talk at Minisymposium for Advances in Optimization and Feasibility Methods for and with Machine Learning, SIAM-Optimization
An Optimal Transport Approach to Continuous Normalizing Flows
Talk at SINE Reading Group, Colorado School of Mines
An Optimal Transport Approach to Continuous Normalizing Flows
Talk at Kernel Club Reading Group, Colorado School of Mines
Adding Value to Hyperspectral Data using Machine Learning
Talk at Center to Advance the Science of Exploration to Reclamation in Mining (CASERM) Meeting
Mineralogy Across Scales - Mapping the Subsurface for Advanced Mineral Exploration and Assessment
publications at GSA 2022 Connects
Adding Value to Hyperspectral Data using Machine Learning
Talk at Center to Advance the Science of Exploration to Reclamation in Mining (CASERM) Meeting
Adding Value to Hyperspectral Data using Machine Learning
Talk at Center to Advance the Science of Exploration to Reclamation in Mining (CASERM) Meeting
Adding Value to Hyperspectral Data using Machine Learning
Talk at Center to Advance the Science of Exploration to Reclamation in Mining (CASERM) Meeting
Adding Value to Hyperspectral Data using Machine Learning
Talk at Center to Advance the Science of Exploration to Reclamation in Mining (CASERM) Meeting
Increasing the Value of Hyperspectral Data Using Advanced Machine Learning Techniques
publications at GSA 2020 Connects Online
Kernel Principal Components Analysis
Talk at Kernel Club Reading Group, Colorado School of Mines