Latent Ordinary Differential Equations for Irregularly-Sampled Time Series

By Yulia Rubanova, T. Chen, David Kristjanson Duvenaud. Neural Information Processing Systems, 2019

912 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/d2eeec5bd43496c968f4b612cc3ede6d85c02a3d

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Learning Differential Equations that are Easy to Solve (2020, methodology): “Code is included for processing the dataset, and links to downloading the data may be found in the code for Rubanova et al. (2019).”
Efficiently Modeling Long Sequences with Structured State Spaces (2021, background): “As a continuous-time model, S4 automatically adapts to data sampled at diﬀerent rates, a challenging setting for time series with a dedicated line of work [10, 35, 37].”
Combining Recurrent, Convolutional, and Continuous-time Models with Linear State-Space Layers (2021, methodology): “Examples include the ODE-RNN [45], GRU-ODE-Bayes [16], and ODE-LSTM [32], which extend adjoint-based neural ODEs to the discrete input setting as an alternative to standard RNNs. Neural Controlled Diﬀerential Equations (NCDE) [31] and Neural Rough Diﬀerential Equations (NRDE) [37] are memory…”
Continuous-in-Depth Neural Networks (2020, background): “Others have rediscovered integration schemes such as the implicit trapezoidal rule, which coincides with the scaled Cayley transform in the context of recurrent neural networks (RNNs) [33].”
CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation (2021, methodology): “10% missing 50% missing 90% missing Latent ODE [35] 0.”
Variable-rate discrete representation learning (2021, background): “Many such variants of recurrent and convolutional neural networks have been explored (Mozer et al., 2017; Neil et al., 2017; Binkowski et al., 2018; Wu et al., 2018b; Rubanova et al., 2019; Shukla & Marlin, 2019).”
Neural Controlled Differential Equations for Irregular Time Series (2020, methodology): “…applies a learnt exponential decay to the hidden state between observations; GRU-ODE [14, 25], which is an ODE analogous to the operation of a GRU and uses X as its input; ODE-RNN [13], which is a GRU-∆ t model which additionally applies a learnt Neural ODE to the hidden state between observations.”
HiPPO: Recurrent Memory with Optimal Polynomial Projections (2020, methodology): “We anticipate that the study of this core problem will be useful in improving a variety of sequence models, and are excited about future work on integrating our memory mechanisms with other models in addition to RNNs.”
Multi-Time Attention Networks for Irregularly Sampled Time Series (2020, methodology): “…progress over the last decade on building and adapting machine learning models that can better capture the structure of irregularly sampled multivariate time series (Li & Marlin, 2015; 2016; Lipton et al., 2016; Futoma et al., 2017; Che et al., 2018; Shukla & Marlin, 2019; Rubanova et al., 2019).”
Diffusion-based Time Series Imputation and Forecasting with Structured State Space Models (2022, methodology): “(3) Finally, there is a collection of approaches relying in modern architectures such as graph neural networks (GRIN)[9], and permutation equivariant networks (NRTSI)[44], self-attention to capture temporal and feature correlations (SAITS) [12], controlled differential equations networks (NeuralCDE)”
How to train your neural ODE (2020, background): “, 2019), in modeling irregular time series (Rubanova et al., 2019), mean field games (Ruthotto et al.”