A scalable end-to-end Gaussian process adapter for irregularly sampled time series classification

By Steven Cheng-Xian Li, Benjamin M. Marlin. 2016

98 citations recorded by the source corpus.

We present a general framework for classification of sparse and irregularly-sampled time series. The properties of such time series can result in substantial uncertainty about the values of the underlying temporal processes, while making the data difficult to deal with using standard classification methods that assume fixed-dimensional feature spaces. To address these challenges, we propose an uncertainty-aware classification framework based on a special computational layer we refer to as the Gaussian process adapter that can connect irregularly sampled time series data to any black-box classifier learnable using gradient descent. We show how to scale up the required computations based on combining the structured kernel interpolation framework and the Lanczos approximation method, and how to discriminatively train the Gaussian process adapter in combination with a number of classifiers end-to-end using backpropagation.

Subfields: Artificial Intelligence, Control and Systems Engineering, Signal Processing

How researchers see this paper

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

In Artificial Intelligence

In Control and Systems Engineering

In Signal Processing

Citations in the corpus

Neural Controlled Differential Equations for Irregular Time Series (2020, background): “For example [6, 7, 8, 9] incorporate exponential decay between observations, [10, 11] hybridise a Gaussian process with traditional neural network models, [12] approximate the underlying continuous-time process, and [13, 14] adapt recurrent neural networks by allowing some hidden state to evolve as…”
Scalable Joint Models for Reliable Uncertainty-Aware Event Prediction (2017, background): “For instance, Li and Marlin [63] proposed a framework for classiﬁcation of (univariate) irregularly sampled time series using GPs.”
An Improved Multi-Output Gaussian Process RNN with Real-Time Validation for Early Sepsis Detection (2017, methodology): “(2016a)) use Long-Short Term Memory (LSTM) RNNs to predict diagnosis codes given physiological time series from the ICU, and (Choi et al. (2016b)) use Gated Recurrent Unit RNNs to predict onset of heart failure using categorical time series of billing codes.”
Learning to Detect Sepsis with a Multitask Gaussian Process RNN Classifier (2017, methodology): “We build off the ideas in (Cheng-Xian Li & Marlin, 2016) to learn a classiﬁer that directly takes the latent function values z i at shared reference time points { } The time series for each encounter i in our data can be represented as an MGP posterior distribution z i ∼ N ( µ z i , Σ z i ; θ ) at…”
Targeting EEG/LFP Synchrony with Neural Nets (2017, methodology): “Following [15], only the expectation of the signal is used (to facilitate fast computation), which is given by X An illustration of the learned new locations is shown under X ∗ in Figure 2.”