skip to main content


The NSF Public Access Repository (NSF-PAR) system and access will be unavailable from 5:00 PM ET until 11:00 PM ET on Friday, June 21 due to maintenance. We apologize for the inconvenience.

This content will become publicly available on August 1, 2024

Title: Self-Recover: Forecasting Block Maxima in Time Series from Predictors with Disparate Temporal Coverage Using Self-Supervised Learning

Forecasting the block maxima of a future time window is a challenging task due to the difficulty in inferring the tail distribution of a target variable. As the historical observations alone may not be sufficient to train robust models to predict the block maxima, domain-driven process models are often available in many scientific domains to supplement the observation data and improve the forecast accuracy. Unfortunately, coupling the historical observations with process model outputs is a challenge due to their disparate temporal coverage. This paper presents Self-Recover, a deep learning framework to predict the block maxima of a time window by employing self-supervised learning to address the varying temporal data coverage problem. Specifically Self-Recover uses a combination of contrastive and generative self-supervised learning schemes along with a denoising autoencoder to impute the missing values. The framework also combines representations of the historical observations with process model outputs via a residual learning approach and learns the generalized extreme value (GEV) distribution characterizing the block maxima values. This enables the framework to reliably estimate the block maxima of each time window along with its confidence interval. Extensive experiments on real-world datasets demonstrate the superiority of Self-Recover compared to other state-of-the-art forecasting methods.

more » « less
Award ID(s):
Author(s) / Creator(s):
; ; ;
Date Published:
Journal Name:
Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence (IJCAI 2023)
Page Range / eLocation ID:
3723 to 3731
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Solar energy is now the cheapest form of electricity in history. Unfortunately, significantly increasing the electric grid's fraction of solar energy remains challenging due to its variability, which makes balancing electricity's supply and demand more difficult. While thermal generators' ramp rate---the maximum rate at which they can change their energy generation---is finite, solar energy's ramp rate is essentially infinite. Thus, accurate near-term solar forecasting, or nowcasting, is important to provide advance warnings to adjust thermal generator output in response to variations in solar generation to ensure a balanced supply and demand. To address the problem, this paper develops a general model for solar nowcasting from abundant and readily available multispectral satellite data using self-supervised learning. Specifically, we develop deep auto-regressive models using convolutional neural networks (CNN) and long short-term memory networks (LSTM) that are globally trained across multiple locations to predict raw future observations of the spatio-temporal spectral data collected by the recently launched GOES-R series of satellites. Our model estimates a location's near-term future solar irradiance based on satellite observations, which we feed to a regression model trained on smaller site-specific solar data to provide near-term solar photovoltaic (PV) forecasts that account for site-specific characteristics. We evaluate our approach for different coverage areas and forecast horizons across 25 solar sites and show that it yields errors close to that of a model using ground-truth observations. 
    more » « less
  2. Electronic health records (EHRs) have been heavily used in modern healthcare systems for recording patients' admission information to health facilities. Many data-driven approaches employ temporal features in EHR for predicting specific diseases, readmission times, and diagnoses of patients. However, most existing predictive models cannot fully utilize EHR data, due to an inherent lack of labels in supervised training for some temporal events. Moreover, it is hard for the existing methods to simultaneously provide generic and personalized interpretability. To address these challenges, we propose Sherbet, a self-supervised graph learning framework with hyperbolic embeddings for temporal health event prediction. We first propose a hyperbolic embedding method with information flow to pretrain medical code representations in a hierarchical structure. We incorporate these pretrained representations into a graph neural network (GNN) to detect disease complications and design a multilevel attention method to compute the contributions of particular diseases and admissions, thus enhancing personalized interpretability. We present a new hierarchy-enhanced historical prediction proxy task in our self-supervised learning framework to fully utilize EHR data and exploit medical domain knowledge. We conduct a comprehensive set of experiments on widely used publicly available EHR datasets to verify the effectiveness of our model. Our results demonstrate the proposed model's strengths in both predictive tasks and interpretable abilities. 
    more » « less
  3. Vedaldi, A. ; Bischof, H. ; Brox, T. ; Frahm, JM. (Ed.)
    The problem of action localization involves locating the action in the video, both over time and spatially in the image. The current dominant approaches use supervised learning to solve this problem. They require large amounts of annotated training data, in the form of frame-level bounding box annotations around the region of interest. In this paper, we present a new approach based on continual learning that uses feature-level predictions for self-supervision. It does not require any training annotations in terms of frame-level bounding boxes. The approach is inspired by cognitive models of visual event perception that propose a prediction-based approach to event understanding. We use a stack of LSTMs coupled with a CNN encoder, along with novel attention mechanisms, to model the events in the video and use this model to predict high-level features for the future frames. The prediction errors are used to learn the parameters of the models continuously. This self-supervised framework is not complicated as other approaches but is very effective in learning robust visual representations for both labeling and localization. It should be noted that the approach outputs in a streaming fashion, requiring only a single pass through the video, making it amenable for real-time processing. We demonstrate this on three datasets - UCF Sports, JHMDB, and THUMOS’13 and show that the proposed approach outperforms weakly-supervised and unsupervised baselines and obtains competitive performance compared to fully supervised baselines. Finally, we show that the proposed framework can generalize to egocentric videos and achieve state-of-the-art results on the unsupervised gaze prediction task. 
    more » « less
  4. Few-shot machine learning attempts to predict outputs given only a very small number of training examples. The key idea behind most few-shot learning approaches is to pre-train the model with a large number of instances from a different but related class of data, classes for which a large number of instances are available for training. Few-shot learning has been most successfully demonstrated for classification problems using Siamese deep learning neural networks. Few-shot learning is less extensively applied to time-series forecasting. Few-shot forecasting is the task of predicting future values of a time-series even when only a small set of historic time-series is available. Few-shot forecasting has applications in domains where a long history of data is not available. This work describes deep neural network architectures for few-shot forecasting. All the architectures use a Siamese twin network approach to learn a difference function between pairs of time-series, rather than directly forecasting based on historical data as seen in traditional forecasting models. The networks are built using Long short-term memory units (LSTM). During forecasting, a model is able to forecast time-series types that were never seen in the training data by using the few available instances of the new time-series type as reference inputs. The proposed architectures are evaluated on Vehicular traffic data collected in California from the Caltrans Performance Measurement System (PeMS). The models were trained with traffic flow data collected at specific locations and then are evaluated by predicting traffic at different locations at different time horizons (0 to 12 hours). The Mean Absolute Error (MAE) was used as the evaluation metric and also as the loss function for training. The proposed architectures show lower prediction error than a baseline nearest neighbor forecast model. The prediction error increases at longer time horizons. 
    more » « less
  5. Spatial prediction is to predict the values of the targeted variable, such as PM2.5 values and temperature, at arbitrary locations based on the collected geospatial data. It greatly affects the key research topics in geoscience in terms of obtaining heterogeneous spatial information (e.g., soil conditions, precipitation rates, wheat yields) for geographic modeling and decision-making at local, regional, and global scales. In-situ data, collected by ground-level in-situ sensors, and remote sensing data, collected by satellite or aircraft, are two important data sources for this task. In-situ data are relatively accurate while sparse and unevenly distributed. Remote sensing data cover large spatial areas but are coarse with low spatiotemporal resolution and prone to interference. How to synergize the complementary strength of these two data types is still a grand challenge. Moreover, it is difficult to model the unknown spatial predictive mapping while handling the trade-off between spatial autocorrelation and heterogeneity. Third, representing spatial relations without substantial information loss is also a critical issue. To address these challenges, we propose a novel Heterogeneous Self-supervised Spatial Prediction (HSSP) framework that synergizes multi-source data by minimizing the inconsistency between in-situ and remote sensing observations. We propose a new deep geometric spatial interpolation model as the prediction backbone that automatically interpolates the values of the targeted variable at unknown locations based on existing observations by taking into account both distance and orientation information. Our proposed interpolator is proven to both be the general form of popular interpolation methods and preserve spatial information. The spatial prediction is enhanced by a novel error-compensation framework to capture the prediction inconsistency due to spatial heterogeneity. Extensive experiments have been conducted on real-world datasets and demonstrated our model’s superiority in performance over state-of-the-art models. 
    more » « less