A Survey on Principles, Models and Methods for Learning from Irregularly Sampled Time Series: From Discretization to Attention and Invariance

Irregularly sampled time series data arise naturally in many application domains including biology, ecology, climate science, astronomy, and health. Such data represent fundamental challenges to many classical models from machine learning and statistics due to the presence of non-uniform intervals between observations. However, there has been significant progress within the machine learning community over the last decade on developing specialized models and architectures for learning from irregularly sampled univariate and multivariate time series data. In this survey, we first describe several axes along which approaches differ including what data representations they are based on, what modeling primitives they leverage to deal with the fundamental problem of irregular sampling, and what inference tasks they are designed to perform. We then survey the recent literature organized primarily along the axis of modeling primitives. We describe approaches based on temporal discretization, interpolation, recurrence, attention, and structural invariance. We discuss similarities and differences between approaches and highlight primary strengths and weaknesses.

Learning from Irregularly-Sampled Time Series: A Missing Data Perspective

Irregularly-sampled time series occur in many domains including healthcare. They can be challenging to model because they do not naturally yield a fixed-dimensional representation as required by many standard machine learning models. In this paper, we consider irregular sampling from the perspective of missing data. We model observed irregularly-sampled time series data as a sequence of index-value pairs sampled from a continuous but unobserved function. We introduce an encoder-decoder framework for learning from such generic indexed sequences. We propose learning methods for this framework based on variational autoencoders and generative adversarial networks. For continuous irregularly-sampled time series, we introduce continuous convolutional layers that can efficiently interface with existing neural network architectures. Experiments show that our models are able to achieve competitive or better classification results on irregularly-sampled multivariate time series compared to recent RNN models while offering significantly faster training times.

URSABench: Comprehensive Benchmarking of Approximate Bayesian Inference Methods for Deep Neural Networks

While deep learning methods continue to improve in predictive accuracy on a wide range of application domains, significant issues remain with other aspects of their performance including their ability to quantify uncertainty and their robustness. Recent advances in approximate Bayesian inference hold significant promise for addressing these concerns, but the computational scalability of these methods can be problematic when applied to large-scale models. In this paper, we describe initial work on the development ofURSABench(the Uncertainty, Robustness, Scalability, and Accu-racy Benchmark), an open-source suite of bench-marking tools for comprehensive assessment of approximate Bayesian inference methods with a focus on deep learning-based classification tasks

Generalized Bayesian Posterior Expectation Distillation for Deep Neural Networks

In this paper, we present a general framework for distilling expectations with respect to the Bayesian posterior distribution of a deep neural network classifier, extending prior work on the Bayesian Dark Knowledge framework. The proposed framework takes as input "teacher" and student model architectures and a general posterior expectation of interest. The distillation method performs an online compression of the selected posterior expectation using iteratively generated Monte Carlo samples. We focus on the posterior predictive distribution and expected entropy as distillation targets. We investigate several aspects of this framework including the impact of uncertainty and the choice of student model architecture. We study methods for student model architecture search from a speed-storage-accuracy perspective and evaluate down-stream tasks leveraging entropy distillation including uncertainty ranking and out-of-distribution detection.

Integrating Physiological Time Series and Clinical Notes with Deep Learning for Improved ICU Mortality Prediction

Intensive Care Unit Electronic Health Records (ICU EHRs) store multimodal data about patients including clinical notes, sparse and irregularly sampled physiological time series, lab results, and more. To date, most methods designed to learn predictive models from ICU EHR data have focused on a single modality. In this paper, we leverage the recently proposed interpolation-prediction deep learning architecture(Shukla and Marlin 2019) as a basis for exploring how physiological time series data and clinical notes can be integrated into a unified mortality prediction model. We study both early and late fusion approaches and demonstrate how the relative predictive value of clinical text and physiological data change over time. Our results show that a late fusion approach can provide a statistically significant improvement in mortality prediction performance over using individual modalities in isolation.

Assessing the Adversarial Robustness of Monte Carlo and Distillation Methods for Deep Bayesian Neural Network Classification

In this paper, we consider the problem of assessing the adversarial robustness of deep neural network models under both Markov chain Monte Carlo (MCMC) and Bayesian Dark Knowledge (BDK) inference approximations. We characterize the robustness of each method to two types of adversarial attacks: the fast gradient sign method (FGSM) and projected gradient descent (PGD). We show that full MCMC-based inference has excellent robustness, significantly outperforming standard point estimation-based learning. On the other hand, BDK provides marginal improvements. As an additional contribution, we present a storage-efficient approach to computing adversarial examples for large Monte Carlo ensembles using both the FGSM and PGD attacks.

Interpolation-Prediction Networks for Irregularly Sampled Time Series

In this paper, we present a new deep learning architecture for addressing the problem of supervised learning with sparse and irregularly sampled multivariate time series. The architecture is based on the use of a semi-parametric interpolation network followed by the application of a prediction network. The interpolation network allows for information to be shared across multiple dimensions of a multivariate time series during the interpolation stage, while any standard deep learning model can be used for the prediction network. This work is motivated by the analysis of physiological time series data in electronic health records, which are sparse, irregularly sampled, and multivariate. We investigate the performance of this architecture on both classification and regression tasks, showing that our approach outperforms a range of baseline and recently proposed models.

Assessing the Robustness of Bayesian Dark Knowledge to Posterior Uncertainty

Bayesian Dark Knowledge is a method for compressing the posterior predictive distribution of a neural network model into a more compact form. Specifically, the method attempts to compress a Monte Carlo approximation to the parameter posterior into a single network representing the posterior predictive distribution. Further, the authors show that this approach is successful in the classification setting using a student network whose architecture matches that of a single network in the teacher ensemble. In this work, we examine the robustness of Bayesian Dark Knowledge to higher levels of posterior uncertainty. We show that using a student network that matches the teacher architecture may fail to yield acceptable performance. We study an approach to close the resulting performance gap by increasing student model capacity.

Integrating Propositional and Relational Label Side Information for Hierarchical Zero-Shot Image Classification

Zero-shot learning (ZSL) is one of the most extreme forms of learning from scarce labeled data. It enables predicting that images belong to classes for which no labeled training instances are available. In this paper, we present a new ZSL framework that leverages both label attribute side information and a semantic label hierarchy. We present two methods, lifted zero-shot prediction and a custom conditional random field (CRF) model, that integrate both forms of side information. We propose benchmark tasks for this framework that focus on making predictions across a range of semantic levels. We show that lifted zero-shot prediction can dramatically outperform baseline methods when making predictions within specified semantic levels, and that the probability distribution provided by the CRF model can be leveraged to yield further performance improvements when making unconstrained predictions over the hierarchy.