davina@zamanzadeh: ~

Hi! I'm Davina.

I'm a machine learning engineer/data scientist.

  • more about_davina.txt

    I am a (San Fernando) Valley girl, born and raised.

    Received my PhD at UCLA in computer science (CS) with an emphasis in machine learning and eHealth.

    Received my BSc at UCSB, also in CS.

    Always down to talk about skiing, martial arts, video games, or sand volleyball.

  • wget my_resume.pdf
  • wget my_cv.pdf
  • more about_experience.txt

    Data Science Intern at Microsoft

    I worked on the Bing Ads (Algorithms and Infrastructure) team during the summer of 2021, developed a semi-novel knowledge distillation technique for image deep neural networks (DNN) in a multi-teacher setting.

    Machine Learning Intern at Pinterest

    I worked on the Search team during the summer of 2020, built ranking model to improve query recommendations.

    Software Engineer Intern at Pinterest

    I worked on the Annotations pipeline as part of the Content/Knowledge team during summer 2017.

    Software Engineer Intern at Impact Radius

    Spent a year at Impact Radius part time during the academic year, full time over summer 2015-2016.

    I worked on improving their metrics and monitoring systems as well as internationalization and localization workflow.

    Software Engineer Intern at GLENWorld

    Developed new features and mini-games for the XPrize Global Learning Competition, part time for a few months during 2015.

    Software Engineer Intern at Datapop

    Worked on internal tooling to optimize company workflow and to resolve conflicts across taxonomies, during summer of 2013 and 2014.

  • more about_research.txt

    Dissertation

    Imputation Is a Hyperparameter: Imputation Deep Learning Model Selection and Evaluation on Large Clinical Datasets
    The framework I built for my work is available on GitHub. More about the dissertation below.

    Researcher @eHealth Research Lab (eR Lab) & Medical Imaging and Informatics Group (MII) @ UCLA

    Working under Majid Sarrafzadeh and Alex Bui to do research in machine learning (ML) with a focus on applications in health and medicine.

    • CURE CKD: I am currently working with nephrologists at UCLA Health to predict rapid kidney function decline in patients with chronic kidney disease.
    • CRRT: I also have another project with nephrologists at UCLA Health on predicting if patients near kidney failure will benefit from a gentler form of dialysis called CRRT.
    • Project REFOCUS: Previously part of a collaboration between UCLA Public Health, Howard University, and the CDC to contruct a racism-aware surveillance system.
    • CD: Previously, I spearheaded a project to tackle clinical questions revolving around Crohn’s Disease (CD) in collaboration with a research group at Cedars Sinai.

    Theoretical sub-fields I'm interested in include imputation and graph representation learning.

    Programming Languages Lab (UCSB)

    Working under Prof. Ben Hardekopf to find an effective comparison metric, based on linear algebra, to compare test sets/fuzzers.

    Assisted writing a paper with Prof. Hardekopf and PhD student Miroslav Gavrilov based on this research.

  • more about_dissertation.txt

    Abstract
    Many real-world datasets suffer from missing data, which can introduce uncertainty into ensuing analyses. To address missing data, researchers have been developing, analyzing, and comparing statistical and machine learning techniques for missing data estimation or imputation. In this context, we built an original framework, Autopopulus, and performed novel analyses that explored predictive pipelines using flexible autoencoder-led imputation. Our work examines autoencoder-led imputation with a deeper regard for the taxonomy of missingness scenarios and mixed feature data of large real-world clinical datasets. In this dissertation we quantify, in a direct manner, the extent to which different methods of imputation affect downstream tasks, and therefore provide rationale for how to choose a solution for a particular dataset and task. We illuminate important decision-making points when assembling a data processing pipeline that handles missing data, while our framework itself allows researchers to apply and compare solutions directly in a unified way for any large dataset. We find that there are different imputation traits under a more granular classification of missingness scenarios, and that trends between imputation performance superiority and predictive performance superiority do not align. Based on our exploration, we believe that the characterization of missingness in the literature must be expanded and that imputing accurately is not always necessary for predicting accurately. We are just beginning to have a clearer view of just how wide the gap is in our understanding and classification of missingness and have hope that this new information will lead to progress in comprehending both the unknown and the unknowable

  • more about_publications.txt
  • more about_awards_and_activities.txt

    Featured Guest on Health and Explainable AI Podcast (03/2023)

    Discussed my experience working on intersectional projects in the clinical domain, and my ideas on the future of responsible AI in health.

    Scientific Python Community Manager

    Help moderate the Discord and the Discuss forum online of the Scientific Python community.

    NIH KUH-ART TL1 Training Grant (2021 - 2023)

    Awarded fellowship to conduct research in the areas relevant to benign nephrology, urology and hematology.

    Podcast Treasurer and Content Advisor (2020 - Present)

    Check us out at @FYInformatics (For Your Informatics). I used run the Twitter account, help create visuals, logo animations, etc. Now I manage the funds and help write and narrate the occasional episode.

    NIH T32 Training Grant (2018 - 2020)

    Awarded fellowship to pursue informatics towards improving health outcomes on clinical data.

    Graduate Dean’s Scholar Award (2018 & 2019)

    Departmental award given to highly recruited students who are deemed to enhance UCLA’s competitiveness.

    Grace Hopper Celebration Scholarship (2017)

    The Computer Science department at UCSB funded some of its students for the first time to attend the Grace Hopper Celebration of Women in Computing.

    Honestly a wonderful event, I am looking forward to being able to attend again in the future.

    KPCB Fellow and Decision Committee (2017)

    I had the honor of being a part of the Kleiner Perkins Fellows program while interning at Pinterest.

    After surviving eliminations out of thousands of applicants, I was able to join 80 other Fellows in many networking events and workshops with technical leaders in Silicon Valley.

    Additionally, I served on the Decision Committee to help pick up potential Fellows for 2018.

    Phi Sigma Rho Chapter Founder

    Phi Sigma Rho is a nationally recognized sorority focused on women in STEM.

    After responding to an an email blast for interest, I helped establish the Alphi Xi chapter of Phi Sigma Rho at UCSB and became part of the founding class in Spring of 2016.

    It wasn't until Fall of 2017 that we became an official chapter, where we all wanted to cry of happiness because our hard work had paid off.

    Through the process of being in charge of designs, involved in planning events, writing bylaws, recruiting, and mentorship, I had an incredible and rewarding experience with other women engineers that I will always remember.

    SB Hacks Organizer

    Starting from my freshman year to my senior year, I watched as SB Hacks grew into something incredible.

    I was involved with logistics (venue, food, etc), design, and getting sponsors throughout my time with the team.

    Thomas Edison Award (2013)

    I participated and won first place in my very first hackathon.

    Out of 15 teams, and hours I don't want to think about, my team won for most creative and useful project for the company.

    Programming Club President and Founder (High School)

    After taking Intro to CS at UCLA after my freshman year, I decided that I alone should not have to suffer the wrath of C++.

    I saw there was no programming clubs at my school, so I made one.

    In some odd foreshadowing of continuing to graduate school, I created lessons plans as well as coding exercises to teach C++ to club members.

  • more credits.txt

    Terminal colors inspired by Solarized theme, or Dracula theme depending on my mood.

    Resume designed and built by Davina Zamanzadeh with XeLaTeX.

    CV design by Alessandro Plasmati, which inspired my design.

    Website designed and built by Davina Zamanzadeh with Stack Overflow.