Skip to yearly menu bar Skip to main content


Interpolation Technique to Speed Up Gradients Propagation in Neural ODEs

Talgat Daulbaev · Alexandr Katrutsa · Larisa Markeeva · Julia Gusak · Andrzej Cichocki · Ivan Oseledets

Poster Session 1 #379


We propose a simple interpolation-based method for the efficient approximation of gradients in neural ODE models. We compare it with reverse dynamic method (known in literature as “adjoint method”) to train neural ODEs on classification, density estimation and inference approximation tasks. We also propose a theoretical justification of our approach using logarithmic norm formalism. As a result, our method allows faster model training than the reverse dynamic method what was confirmed and validated by extensive numerical experiments for several standard benchmarks.

Chat is not available.