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 …
Robust heterogeneous federated learning under data corruption
Abstract Model heterogeneous federated learning is a realistic and challenging problem.
However, due to the limitations of data collection, storage, and transmission conditions, as …
However, due to the limitations of data collection, storage, and transmission conditions, as …
On robustness of neural ordinary differential equations
Neural ordinary differential equations (ODEs) have been attracting increasing attention in
various research domains recently. There have been some works studying optimization …
various research domains recently. There have been some works studying optimization …
Neural stochastic pdes: Resolution-invariant learning of continuous spatiotemporal dynamics
C Salvi, M Lemercier… - Advances in Neural …, 2022 - proceedings.neurips.cc
Stochastic partial differential equations (SPDEs) are the mathematical tool of choice for
modelling spatiotemporal PDE-dynamics under the influence of randomness. Based on the …
modelling spatiotemporal PDE-dynamics under the influence of randomness. Based on the …
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 modular simulations for homogeneous systems
Complex systems are often decomposed into modular subsystems for engineering
tractability. Although various equation based white-box modeling techniques make use of …
tractability. Although various equation based white-box modeling techniques make use of …
Continuous-in-depth neural networks
Recent work has attempted to interpret residual networks (ResNets) as one step of a forward
Euler discretization of an ordinary differential equation, focusing mainly on syntactic …
Euler discretization of an ordinary differential equation, focusing mainly on syntactic …
Learning stochastic dynamics with statistics-informed neural network
We introduce a machine-learning framework named statistics-informed neural network
(SINN) for learning stochastic dynamics from data. This new architecture was theoretically …
(SINN) for learning stochastic dynamics from data. This new architecture was theoretically …
Certified robustness for deep equilibrium models via interval bound propagation
Deep equilibrium layers (DEQs) have demonstrated promising performance and are
competitive with standard explicit models on many benchmarks. However, little is known …
competitive with standard explicit models on many benchmarks. However, little is known …
Look beneath the surface: Exploiting fundamental symmetry for sample-efficient offline rl
Offline reinforcement learning (RL) offers an appealing approach to real-world tasks by
learning policies from pre-collected datasets without interacting with the environment …
learning policies from pre-collected datasets without interacting with the environment …