Roman Yakunin

Machine Learning Research Intern at GaTech EIC Lab
  • Claim this Profile
Contact Information
us****@****om
(386) 825-5501
Location
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.

Credentials

  • Deep Learning Specialization
    Coursera
    Jun, 2021
    - Nov, 2024

Experience

    • United States
    • Research Services
    • 1 - 100 Employee
    • Machine Learning Research Intern
      • Sep 2023 - Present

      Researching compression methods (such as low-bit quantization, pruning, and knowledge distillation) for large language models like Llama 2 for the purpose of reducing model size, accelerating inference, and improving energy efficiency Researching compression methods (such as low-bit quantization, pruning, and knowledge distillation) for large language models like Llama 2 for the purpose of reducing model size, accelerating inference, and improving energy efficiency

  • UC Irvine Biorobotics Laboratory
    • Irvine, California, United States
    • Machine Learning Research Intern
      • Jun 2021 - Jun 2023

      • Built machine learning models to detect hand opening & closing based off wrist accelerometer and gyroscope measurements for the purpose of passively tracking and promoting the recovery of stroke patients with impaired arm control • Designed data processing pipeline that converts files of time-series sensor data to shards of Box-Cox normalized spectrogram samples • Created system for automatically organizing preprocessed data and results for dozens of tests • Compared the performances of 7 combinations of data processing and machine learning methods (such as KNN, SVM, Dense NN, CNN) across several datasets • Designed a custom convolutional neural network, yielding the most accurate approach for all tests • Produced a method strongly correlated to established HAND algorithm approach yet more practical, paving way for potential deployment on various existing smartwatches for healthcare monitoring • As second co-author, aided in writing and editing paper detailing this work, which was published in the peer-reviewed journal Sensors Show less

Education

  • Georgia Institute of Technology
    Bachelor of Science - BS, Computer Science
    2023 - 2025
  • College of the Canyons
    Associate Degrees for Transfer in Mathematics and Physics, GPA: 3.99
    2018 - 2022
  • Academy of the Canyons
    High School Diploma, W/UW GPA: 4.69/3.97
    2018 - 2022

Community

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