Gorkem Ates
Development
FL, United States
Skills
Machine Learning (ML)
About
Gorkem Can Ates's skills align with IT R&D Professionals (Information and Communication Technology). Gorkem also has skills associated with Financial Experts (Insurance and Finance). Gorkem Can Ates has 6 years of work experience.
View more
Work Experience
Machine Learning Engineer (PhD Intern)
Regeneron Pharmaceuticals
June 2023 - September 2023
- Tarrytown, NY, USA • Explored self-supervised computer vision frameworks such as DINO and masked au- Jun. 2023 - Sept. 2023 toencoders. • Developed a self-supervised learning framework using DINO and hierarchical vision transformers for gigapixel-level digital pathology images. • The developed framework provides accurate hierarchical segmentation of nuclei, cells, etc., without the need for any labels.
Graduate Research Assistant
TOBB University of Economics and Technology
September 2020 - August 2021
- Ankara, Turkey • Developed an advanced convolutional encoder-decoder segmentation network to de- Sept. 2020 - Aug. 2021 tect stroke lesion segmentation using brain CT images for Teknofest Artificial Intelligence in Health Competition • Proposed a novel two-stage convolutional encoder-decoder network for structural topology optimization. Proposed two-stage model advanced the state-of-the-art prediction accuracy in the literature by 3.5%. • Developed a VGG-Unet based encoder-decoder network using transfer learning for structural topology optimization.Proposed model with the transfer learning framework improved the prediction accuracy by 0.13% while reducing the training time by nearly 7%. • Developed a Deep Neural Network to predict the statistical parameters of the effective strut diameters to account for the geometric uncertainties introduced by additive manufacturing.
Research and Developement Engineer, Part-time
Roketsan Missiles Inc
May 2018 - May 2020
- Ankara, Turkey • Roketsan Missiles. is the leading institution in Turkey for designing, developing and May. 2018 - May. 2020 manufacturing rockets and missiles. 1 year 7 months • Developed non-linear system identification for non-linear flight dynamic models. • Designed mathematical guidance algorithms for missiles.
Graduate Research Assistant
University of Miami
August 2021 - Present
- Coral Gables, FL, USA • Developed a real-time and robust CNN-based eye-tracking model using quantization- Aug. 2021 - Current aware training. (90.1% speed improvement with similar accuracy compared to traditional implementations) • Developed an end-to-end computer vision trainer framework from scratch for detection and segmentation applications using Pytorch. • Developed a light-weight dual cross-attention mechanism to improve the segmentation performance for medical image segmentation problems. • Working towards the development of supervised and self-supervised Vision Transformers for the diagnosis and prognosis of HCC using liver CT images.
Education
University of Miami
PhD
August 2021 - May 2024
TOBB University of Economics and Technology
MS
September 2020 - August 2021
