Valentin Anklin

Master's Thesis Student

IBM•  September 2020 - March 2021

• Development of a state-of-the-art Graph Neural Network system for cancer segmentation of histopathological images. • Contributing to the basis of the open-source Python library HistoCartography (Guillaume Jaume, Pushpak Pati, Valentin Anklin, Antonio Foncubierta, Maria Gabrani “HistoCartography: A Toolkit for Graph Analytics in Digital Pathology” arXiv PrePrint, arXiv: 2107.10073 (2021)) • Work published at MICCAI 2021 (Valentin Anklin et al. "Learning Whole-Slide Segmentation from Inexact and Incomplete Labels using Tissue Graphs." MICCAI, arXiv:2103.03129 (2021).)

Teaching Assistant

ETH Zürich•  September 2017 - December 2020

• Planning and coordination as head teaching assistant for the course Robustness of Deep Neural Networks • Organization and Implementation of a Capture the Flag (CTF) Competition on anonymity with mix networks in the Information Security Group • Preparing and holding exercise sessions for ~40 students for Formal Methods and Functional Programming and Numerical Methods for CSE

Machine Learning Engineer, Intern

Visium SA•  February 2020 - July 2020

• Developing a state-of-the-art Computer Vision model for a pharma client • Building and integrating an end-to-end Natural Language Processing pipeline for a recommendation system. • Leading initiatives for reproducible data pipelines, improved knowledge transfer between projects and better code quality guidelines

ETH Zurich - Swiss Federal Institute of Technology, Zurich

Computer Science, MS•  September 2018 - March 2021

• Focus on Machine Learning with various projects: 3D human pose estimation from a single RGB image, road segmentation of satellite images, scalable genetic algorithm, Deep Reinforcement Learning for battleships game. • Overall grade point average 5.61/6.0

ETH Zurich - Swiss Federal Institute of Technology, Zurich

Computer Science, BS•  September 2015 - August 2018

• Thesis on building a novel rapid 3D fabrication pipeline (in C++) at the Interactive Geometry Lab. • Overall grade point average 5.45/6.0