Andrew Cutler

Principal Component Analyst at REACT Neuro
  • Claim this Profile
Contact Information
Location
Boston, Massachusetts, United States, US

Topline Score

Topline score feature will be out soon.

Bio

Generated by
Topline AI

You need to have a working account to view this content.
You need to have a working account to view this content.

Experience

    • United States
    • Wellness and Fitness Services
    • 1 - 100 Employee
    • Principal Component Analyst
      • Apr 2021 - Present

      Using statistics to detect Alzheimer's, tremors, and concussions from tests performed in VR. Using statistics to detect Alzheimer's, tremors, and concussions from tests performed in VR.

    • United States
    • Higher Education
    • 700 & Above Employee
    • Ph.D student
      • Aug 2015 - Jan 2021

      Predict personality from Facebook Statuses. Using models of increasing complexity I increased accuracy from previous standard of r = 0 to: 0.18 ->0.27 -> 0.44. Similar results on other traits such as impulsivity, IQ, and political preference. Models used were BoW, RoBERTa, and RoBERTa + multi-task learning.Also made contributions to our understanding of personality structure by deriving it from language models (unsupervised learning problem). The results are parsimoneous, robust to language (english vs spanish), and align with theories of personality development and evolutionary psychology.

    • Netherlands
    • Non-profit Organization Management
    • 1 - 100 Employee
    • Junior Researcher
      • May 2018 - Sep 2018

      I updated their k-nearest neighbors flood prediction model to be a recurrent neural network. In addition to upstream flow data, I took it upon myself to incorporate a daily map of a) precipitation b) potential evapotranspiration and c) temperature from satellite data. I learned many things about communicating models to decision makers, dealing with missing data, and the benefits of looking for new streams of data. I updated their k-nearest neighbors flood prediction model to be a recurrent neural network. In addition to upstream flow data, I took it upon myself to incorporate a daily map of a) precipitation b) potential evapotranspiration and c) temperature from satellite data. I learned many things about communicating models to decision makers, dealing with missing data, and the benefits of looking for new streams of data.

    • United States
    • Software Development
    • 1 - 100 Employee
    • Machine Learning Engineer
      • 2017 - 2018

      Built a system that could classify reviews of employees by topic (communication skills vs technical skills) and sentiment. This involved the collection of data, training models, and working with the web development team to get them operating in real time. Built a system that could classify reviews of employees by topic (communication skills vs technical skills) and sentiment. This involved the collection of data, training models, and working with the web development team to get them operating in real time.

    • United States
    • Higher Education
    • 700 & Above Employee
    • Research Assistant
      • Sep 2012 - Aug 2015

      Clean room work. I fabricated a device for in situ measurements of Li-ion battery cathodes. Clean room work. I fabricated a device for in situ measurements of Li-ion battery cathodes.

Education

  • Boston University
    Doctor of Philosophy - PhD, Electrical Engineering (information data systems)
    2015 - 2021
  • Brigham Young University
    Bachelor of Science (BS), Electrical Engineering
    2009 - 2015

Community

You need to have a working account to view this content. Click here to join now