Mizu Nishikawa-Toomey

About Me

I am a PhD student at Mila - Montreal Institute of Learning Algorithms,
under the supervision of Laurent Charlin and Dhanya Sridhar.
Previously, I was a masters student, then a RA at the OATML group at the University of Oxford,
supervised by Yarin Gal.
Before that, I worked at Oxford Nanopore Technologies, using machine learning
for Nanopore DNA sequencing.
My bachelors was in physics which I undertook at University College London.

Research Interests

I am interested in causal structure learning, decision making under uncertainty
and in general, asking -
what real world interventions are most useful for our ML model and the questions posed to it.

Papers

AAAI 2023 Workshop on Graphs and Complex structures for Learning and Reasoning

Bayesian learning of Causal Structure and Mechanisms
with GFlowNets and Variational Bayes

Neurips 2020 Workshop on Bayesian Deep Learning

Semi-supervised Learning of Galaxy Morphology using
Equivariant Transformer Variational Autoencoders