Scalable gradients for stochastic differential equations
The adjoint sensitivity method scalably computes gradients of solutions to ordinary
differential equations. We generalize this method to stochastic differential equations …
differential equations. We generalize this method to stochastic differential equations …
Ode2vae: Deep generative second order odes with bayesian neural networks
C Yildiz, M Heinonen… - Advances in Neural …, 2019 - proceedings.neurips.cc
Abstract We present Ordinary Differential Equation Variational Auto-Encoder (ODE2VAE), a
latent second order ODE model for high-dimensional sequential data. Leveraging the …
latent second order ODE model for high-dimensional sequential data. Leveraging the …
Macroscopic traffic flow modeling with physics regularized Gaussian process: A new insight into machine learning applications in transportation
Despite the wide implementation of machine learning (ML) technique in traffic flow modeling
recently, those data-driven approaches often fall short of accuracy in the cases with a small …
recently, those data-driven approaches often fall short of accuracy in the cases with a small …
Continuous-time model-based reinforcement learning
C Yildiz, M Heinonen… - … Conference on Machine …, 2021 - proceedings.mlr.press
Abstract Model-based reinforcement learning (MBRL) approaches rely on discrete-time state
transition models whereas physical systems and the vast majority of control tasks operate in …
transition models whereas physical systems and the vast majority of control tasks operate in …
Scalable gradients and variational inference for stochastic differential equations
We derive reverse-mode (or adjoint) automatic differentiation for solutions of stochastic
differential equations (SDEs), allowing time-efficient and constant-memory computation of …
differential equations (SDEs), allowing time-efficient and constant-memory computation of …
Learning interacting dynamical systems with latent gaussian process odes
Ç Yıldız, M Kandemir… - Advances in Neural …, 2022 - proceedings.neurips.cc
We study uncertainty-aware modeling of continuous-time dynamics of interacting objects.
We introduce a new model that decomposes independent dynamics of single objects …
We introduce a new model that decomposes independent dynamics of single objects …
[HTML][HTML] Continual learning from demonstration of robotics skills
Methods for teaching motion skills to robots focus on training for a single skill at a time.
Robots capable of learning from demonstration can considerably benefit from the added …
Robots capable of learning from demonstration can considerably benefit from the added …
FAIR: Fair collaborative active learning with individual rationality for scientific discovery
Scientific discovery aims to find new patterns and test specific hypotheses by analysing
large-scale experimental data. However, various practical limitations (eg, high experimental …
large-scale experimental data. However, various practical limitations (eg, high experimental …
Learning efficient and robust ordinary differential equations via invertible neural networks
Advances in differentiable numerical integrators have enabled the use of gradient descent
techniques to learn ordinary differential equations (ODEs), where a flexible function …
techniques to learn ordinary differential equations (ODEs), where a flexible function …
Neural graphical modelling in continuous-time: consistency guarantees and algorithms
A Bellot, K Branson, M van der Schaar - arXiv preprint arXiv:2105.02522, 2021 - arxiv.org
The discovery of structure from time series data is a key problem in fields of study working
with complex systems. Most identifiability results and learning algorithms assume the …
with complex systems. Most identifiability results and learning algorithms assume the …