Mykola Liashuha

Deep Learning Researcher at CEA-List
  • Claim this Profile
Contact Information
Location
Orsay, Île-de-France, France, FR

Topline Score

Bio

Generated by
Topline AI

0

/5.0
/ Based on 0 ratings
  • (0)
  • (0)
  • (0)
  • (0)
  • (0)

Filter reviews by:

No reviews to display There are currently no reviews available.

0

/5.0
/ Based on 0 ratings
  • (0)
  • (0)
  • (0)
  • (0)
  • (0)

Filter reviews by:

No reviews to display There are currently no reviews available.
You need to have a working account to view this content. Click here to join now

Credentials

  • Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization
    Coursera
    May, 2020
    - Sep, 2024
  • Autonomous Cars: Deep Learning and Computer Vision
    Udemy
    Dec, 2019
    - Sep, 2024
  • IELTS Academic 7.0
    IELTS Official
    Oct, 2019
    - Sep, 2024

Experience

    • France
    • Think Tanks
    • 100 - 200 Employee
    • Deep Learning Researcher
      • Oct 2022 - Present

      Moonshot project: Training and integrating multi-task vision robotic-arm model using RL and DAgger Imitation Learning

    • Machine Learning Research Intern
      • Mar 2022 - Sep 2022

      Done internship at CEA LIST, LVA lab: "Action prediction in football videos with vision transformers" Key idea was to predict what kind of action happens at every second in soccer videos. It was done by doing pre-training of the transformer and fine-tuninng it on SoccerNet Action Spotting Dataset

    • France
    • Higher Education
    • 700 & Above Employee
    • Deep Learning Research Intern
      • May 2021 - Aug 2021

      Detection of Metastatic Breast Cancer in Sentinel Lymph Nodes at MICS Laboratory Detection of Metastatic Breast Cancer in Sentinel Lymph Nodes at MICS Laboratory

    • France
    • Research Services
    • 700 & Above Employee
    • Machine Learning Intern (Travail d'Etude et de Recherche)
      • Nov 2020 - Feb 2021

      Use of Graph Neural Networks for gas-liquid interface reconstruction in CFD numerical simulations. Performed experiments with "Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering", "Inductive Representation Learning on Large Graphs" and "PointCNN: Convolution On X-Transformed Points papers". Use of Graph Neural Networks for gas-liquid interface reconstruction in CFD numerical simulations. Performed experiments with "Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering", "Inductive Representation Learning on Large Graphs" and "PointCNN: Convolution On X-Transformed Points papers".

    • Machine Learning Engineer
      • Aug 2019 - Aug 2020

      Implementing autonomous grocery shop with a similar idea to Amazon Go Using multi-camera view for 3D human pose tracking Involved in instance segmentation and background subtraction tasks Detecting events using signal from weight sensors Implementing autonomous grocery shop with a similar idea to Amazon Go Using multi-camera view for 3D human pose tracking Involved in instance segmentation and background subtraction tasks Detecting events using signal from weight sensors

    • Ukraine
    • IT Services and IT Consulting
    • 1 - 100 Employee
    • Data Scientist
      • Jun 2019 - Aug 2019

      Developing automatic system for product merchandising using AI Detecting products on grocery store shelves with Mask-RCNN Developing automatic system for product merchandising using AI Detecting products on grocery store shelves with Mask-RCNN

    • IT Services and IT Consulting
    • 1 - 100 Employee
    • ASP.NET Student
      • Feb 2018 - Aug 2018

Education

  • Université Paris-Saclay
    M2 Artificial Intelligence, Artificial Intelligence
    2021 - 2022
  • Université Paris-Saclay
    M1 Artificial intelligence, Artificial Intelligence
    2020 - 2021
  • University of Zagreb/Sveuciliste u Zagrebu
    Erasmus+ exchange semester, Computer Science
    2018 - 2019
  • Taras Shevchenko National University of Kyiv
    Bachelor's degree, Software Engineering
    2016 - 2020

Community

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