Open PhD position: Deep learning with differential equations
Supervisor: Academy Research Fellow Markus Heinonen (Aalto University, Finland)
We are looking for an exceptional and motivated phd student to push the boundaries of deep learning with differential equations. We aim at developing state-of-the-art neural and Bayesian differential equation models to open new avenues in predictive and generative settings. Possible topics range from developing interpretable and more efficient ODEs for supervised tasks, to modelling distributional structures with ODEs and PDEs, to adversarial or robust ODEs.
The work continues on the foundations of our NIPS'19 and AISTATS'19 publications "ODE2VAE" and "Differential Gaussian processes".
The work will be done in close collaboration with the supervisor and and with several other researchers studying complementary topics at Aalto, and internationally. Doctoral students are encouraged to make visits to collaboration sites. Candidates are expected to have completed a Masters degree, and have familiarity with machine learning.