Ladislav Rampášek

Research Scientist, Isomorphic Labs  •  Previously: Mila + UdeM / UofT

Ladislav-Mila2022.jpg

Previously, I was a postdoctoral fellow in the group of Guy Wolf (UdeM), investigating questions at the intersection of graph representation learning, graph signal processing, and geometric deep learning.

I completed my PhD at the University of Toronto in Anna Goldenberg’s group, where I developed latent-variable models for cancer treatment outcome prediction and drug perturbation effect modelling.

In fall 2018 I interned in Google Accelerated Science team (Mountain View, CA), where I developed machine learning models for predicting drug effects on cell phenotype from the drug molecular structure.

In 2016 I interned in Atomwise (San Francisco, CA), where I worked on 3D convolutional neural nets to predict small molecule binding to protein targets and for scoring of the 3D binding poses.

My CV can be found here.

selected publications

  1. NeurIPS
    Recipe for a general, powerful, scalable graph transformer
    Ladislav Rampášek, Mikhail Galkin, Vijay Prakash Dwivedi, Anh Tuan Luu, Guy Wolf, and Dominique Beaini
    In 36th Conference on Neural Information Processing Systems (NeurIPS), 2022
    arXiv Blog Code
  2. arXiv
    GPS++: Reviving the Art of Message Passing for Molecular Property Prediction
    Dominic Masters, Josef Dean, Kerstin Klaser, Zhiyi Li, Sam Maddrell-Mander, Adam Sanders, Hatem Helal, and 5 more authors
    arXiv:2302.02947, 2023
    arXiv Code
  3. arXiv
    Attending to Graph Transformers
    Luis Müller, Mikhail Galkin, Christopher Morris, and Ladislav Rampášek
    arXiv:2302.04181, 2023
    arXiv Code