Welcome to my online CV. Links to relevant literature related to my history can be found here. As well, this will eventually become a place to blog about bioengineering, machine learning, and how to write stellar SBIR grants, with an emphasis on the former two. Seeking academic and industry opportunities in computational biology, machine learning research, statistics, and the intersection of the three. Follow me on twitter!

Education

Ph.D. Candidate in Biomedical Engineering University of California, Irvine (2019-)

M.Res. in Bioengineering Imperial College London (2015-2016)

B.S./M.Eng. in Biomedical Engineering, minor in Electrical Engineering Texas A&M University (2010-2015)

Publications

2024

V. D. Zaballa, E. E. Hui, “Reducing Uncertainty Through Mutual Information in Structural and Systems Biology.” ICML 2024 Machine Learning for Life and Material Science, July 2024 (In Review)

Combined data from protein structure predictions and systems biology models to improve systems biology model predictions and support new structural biology hypotheses about how surface receptor proteins may interact.

2023

V. D. Zaballa, E. E. Hui, “Approximation of Intractable Likelihood Functions in Systems Biology via Normalizing Flows.” NeurIPS 2023 Generative AI and Biology Workshop, December 2023 Paper

Showcased scalable Likelihood-Free Inference via Normalizing Flows in biological datasets with 940 dimensions. Demonstrated utility in model selection scenario where two biological models were compared using the trained flows.

V. D. Zaballa, E. E. Hui, “Stochastic Gradient Bayesian Optimal Experimental Designs for Simulation-based Inference.” ICML 2023 Differentiable Everything Workshop, July 2023 Paper

Demonstrated a novel approach for estimating Mutual Information in Simulation-based Inference, enabling amortized inference of intractable likelihood that showcased a theoretical advancement and practical feasibility in high-dimensional optimization compared to Bayesian optimization.

2022

V. D. Zaballa, E. E. Hui, “An Optimal Likelihood Free Method for Biological Model Selection.” ICML 2022 Workshop on Computational Biology, July 2022 paper

Created new method for automated biological model selection that improved convergence to a correct model with 166-298% better accuracy over baseline methods with a fixed budget of experiments.

2021

V. D. Zaballa, E. E. Hui, “Optimal Design of Experiments for Simulation-Based Inference of Mechanistic Acyclic Biological Networks.” NeurIPs Workshop on Learning Meaningful Representations of Life, Dec. 2021 paper video

Improved average experimental design efficiency in systems biology over baseline standards by 225% using Bayesian deep learning and physics-based math models.

2015

V. Zaballa, D. Friedrichs, “A Novel Tissue Welding Device using Low-Voltage Far-Field Coaxial Electrospinning.” ASME Design of Medical Devices Conference Journal, Jan. 2015 paper

Invited Talks & Presentations

“Optimal Design of Experiments for Simulation-Based Inference of Mechanistic Acyclic Biological Networks.” Lightning Talk at SoCalSysBio 2022 on April 2, 2022, Los Angeles, CA.

Industry Experience

Fannin Innovation Studio Houston, Texas, Entrepreneur Fellow, June 2018 – Aug. 2019

Managed commercialization and R&D of pioneering fetal surgery device, resulting in award of $2 million SBIR Phase II Commercialization and Product Development Grant. press release
Engineering manager for novel ELISA device, directing two hardware and one software engineer, resulting in a hardware design freeze and new quality control computer vision software. company website

Drip, LLC. Colorado Springs, Colorado, Data Scientist, April 2017 – October 2017

Developed analytics and machine learning algorithms to draw insights from Colorado water data records within a Software as a Service (SaaS) application. Go Code Colorado 2017 Hackathon winning team

Cambiando Paradigmas La Paz, Bolivia, Data Analyst, Jan. 2017 – June 2018

Clustered & visualized energy data to show 25% of buildings accounted for 70% of energy consumption
Improved returns invested in renewable energy programs by 4.5x compared to previous investment portfolio using a random forest machine learning model and knapsack optimization

Imperial College London London, United Kingdom, Research Assistant & Whitaker Fellow, Oct. 2015 – Oct. 2016

Design of a novel, patient-specific, orthopedic implant to reduce strain experienced by bone
Created an optimized design that reduced strain by 90%, measured by finite element analysis

Teaching & Professional Service

Contributor, Probabilistic Machine Learning: Advanced Topics by Kevin Murphy Contributed normalizing flow JAX examples in Google Colab. Figure Notebook1 Notebook2

University of California, Irvine June 2022, TA, Cancer Systems Biology Short Course: Helped students run end-to-end python gene expression processing notebooks in Google Colab.

Reviewer, AISTATS 2022

Imperial College London Oct. 2015 – Dec. 2015, TA, Maths 2: Multivariable calculus and differential equations.

Awards & Fellowships

National Institutes of Health (NIH) National Institute of General Medical Sciences (NIGMS) National Research Service Award (NRSA) F31 Fellowship (2021-2024)
Google Computer Science Research Mentorship Program Awardee (2021)
Go Code Colorado Hackathon Winning Team (2017)
Whitaker International Fellowship (2015-2016)
Imperial College London Postgraduate Award for Outstanding Students (2015)
Design of Medical Devices 3-in-5 Competition Winning Presentation, press release (2015)