More Like this

1. TransEHR: Self-Supervised Transformer for Clinical Time Series Data

   Xu, Yanbo; Xu, Shangqing; Ramprassad, Manav; Tumanov, Alexey; Zhang, Chao (December 2023, Proceedings of Machine Learning Research)

   Deep neural networks, including the Transformer architecture, have achieved remarkable performance in various time series tasks. However, their effectiveness in handling clinical time series data is hindered by specific challenges: 1) Sparse event sequences collected asynchronously with multivariate time series, and 2) Limited availability of labeled data. To address these challenges, we propose Our code is available at https://github.com/SigmaTsing/TransEHR.git . , a self-supervised Transformer model designed to encode multi-sourced asynchronous sequential data, such as structured Electronic Health Records (EHRs), efficiently. We introduce three pretext tasks for pre-training the Transformer model, utilizing large amounts of unlabeled structured EHR data, followed by fine-tuning on downstream prediction tasks using the limited labeled data. Through extensive experiments on three real-world health datasets, we demonstrate that our model achieves state-of-the-art performance on benchmark clinical tasks, including in-hospital mortality classification, phenotyping, and length-of-stay prediction. Our findings highlight the efficacy of in effectively addressing the challenges associated with clinical time series data, thus contributing to advancements in healthcare analytics. 

   more » « less
2. Benchmarking Deep Learning Interpretability in Time Series Predictions

   Ismail, A. A. (January 2020, 34th Conference on Neural Information Processing Systems (NeurIPS 2020))

   null (Ed.)

   Saliency methods are used extensively to highlight the importance of input features in model predictions. These methods are mostly used in vision and language tasks, and their applications to time series data is relatively unexplored. In this paper, we set out to extensively compare the performance of various saliency-based interpretability methods across diverse neural architectures, including Recurrent Neural Network, Temporal Convolutional Networks, and Transformers in a new benchmark † of synthetic time series data. We propose and report multiple metrics to empirically evaluate the performance of saliency methods for detecting feature importance over time using both precision (i.e., whether identified features contain meaningful signals) and recall (i.e., the number of features with signal identified as important). Through several experiments, we show that (i) in general, network architectures and saliency methods fail to reliably and accurately identify feature importance over time in time series data, (ii) this failure is mainly due to the conflation of time and feature domains, and (iii) the quality of saliency maps can be improved substantially by using our proposed two-step temporal saliency rescaling (TSR) approach that first calculates the importance of each time step before calculating the importance of each feature at a time step. 

   more » « less
3. Benchmarking Deep Learning Interpretability in Time Series Predictions

   Ismail, Aya; Gunady, Mohamed; Bravo, Héctor; Feizi, Soheil (January 2020, Advances in Neural Information Processing Systems Foundation (NeurIPS))

   null (Ed.)

   Saliency methods are used extensively to highlight the importance of input features in model predictions. These methods are mostly used in vision and language tasks, and their applications to time series data is relatively unexplored. In this paper, we set out to extensively compare the performance of various saliency-based interpretability methods across diverse neural architectures, including Recurrent Neural Network, Temporal Convolutional Networks, and Transformers in a new benchmark of synthetic time series data. We propose and report multiple metrics to empirically evaluate the performance of saliency methods for detecting feature importance over time using both precision (i.e., whether identified features contain meaningful signals) and recall (i.e., the number of features with signal identified as important). Through several experiments, we show that (i) in general, network architectures and saliency methods fail to reliably and accurately identify feature importance over time in time series data, (ii) this failure is mainly due to the conflation of time and feature domains, and (iii) the quality of saliency maps can be improved substantially by using our proposed two-step temporal saliency rescaling (TSR) approach that first calculates the importance of each time step before calculating the importance of each feature at a time step. 

   more » « less
4. Capturing Temporal Components for Time Series Classification

   Vavilthota, Venkata Ragavendra; Ramanathan, Ranjith; Aakur, Sathyanarayanan N (September 2024, Springer)

   Analyzing sequential data is crucial in many domains, particularly due to the abundance of data collected from the Internet of Things paradigm. Time series classification, the task of categorizing sequential data, has gained prominence, with machine learning approaches demonstrating remarkable performance on public benchmark datasets. However, progress has primarily been in designing architectures for learning representations from raw data at fixed (or ideal) time scales, which can fail to generalize to longer sequences. This work introduces a \textit{compositional representation learning} approach trained on statistically coherent components extracted from sequential data. Based on a multi-scale change space, an unsupervised approach is proposed to segment the sequential data into chunks with similar statistical properties. A sequence-based encoder model is trained in a multi-task setting to learn compositional representations from these temporal components for time series classification. We demonstrate its effectiveness through extensive experiments on publicly available time series classification benchmarks. Evaluating the coherence of segmented components shows its competitive performance on the unsupervised segmentation task. 

   more » « less
5. Poster: Comparative Study of Transformer Models on a Large Multivariate Time Series HAR Dataset

   Kwon, Hyungtaek; Xie, Zongxing; Liu, Mengjing; Ye, Fan (June 2024, IEEE/ACM international conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE))

   In Activities of Daily Living (ADL) research, which has gained prominence due to the burgeoning aging population, the challenge of acquiring sufficient ground truth data for model training is a significant bottleneck. This obstacle necessitates a pivot towards unsupervised representation learning methodologies, which do not require many labeled datasets. The existing research focused on the tradeoff between the fully supervised model and the unsupervised pre-trained model and found that the unsupervised version outperformed in most cases. However, their investigation did not use large enough Human Activity Recognition (HAR) datasets, both datasets resulting in 3 dimensions. This poster extends the investigation by employing a large multivariate time series HAR dataset and experimenting with the models with different combinations of critical training parameters such as batch size and learning rate to observe the performance tradeoff. Our findings reveal that the pre-trained model is comparable to the fully supervised classification with a larger multivariate time series HAR dataset. This discovery underscores the potential of unsupervised representation learning in ADL extractions and highlights the importance of model configuration in optimizing performance. 

   more » « less