MetaTP

2021 ◽  
Vol 5 (3) ◽  
pp. 1-28
Author(s):  
Weida Zhong ◽  
Qiuling Suo ◽  
Abhishek Gupta ◽  
Xiaowei Jia ◽  
Chunming Qiao ◽  
...  
Keyword(s):  
Time Series ◽  
Large Scale ◽  
Multivariate Time Series ◽  
Modern Society ◽  
Training Data ◽  
Traffic Prediction ◽  
Temporal Prediction ◽  
Reference Space ◽  
Meta Learning ◽  
Real World Datasets

With the popularity of smartphones, large-scale road sensing data is being collected to perform traffic prediction, which is an important task in modern society. Due to the nature of the roving sensors on smartphones, the collected traffic data which is in the form of multivariate time series, is often temporally sparse and unevenly distributed across regions. Moreover, different regions can have different traffic patterns, which makes it challenging to adapt models learned from regions with sufficient training data to target regions. Given that many regions may have very sparse data, it is also impossible to build individual models for each region separately. In this paper, we propose a meta-learning based framework named MetaTP to overcome these challenges. MetaTP has two key parts, i.e., basic traffic prediction network (base model) and meta-knowledge transfer. In base model, a two-layer interpolation network is employed to map original time series onto uniformly-spaced reference time points, so that temporal prediction can be effectively performed in the reference space. The meta-learning framework is employed to transfer knowledge from source regions with a large amount of data to target regions with a few data examples via fast adaptation, in order to improve model generalizability on target regions. Moreover, we use two memory networks to capture the global patterns of spatial and temporal information across regions. We evaluate the proposed framework on two real-world datasets, and experimental results show the effectiveness of the proposed framework.

scholarly journals Joint Modeling of Local and Global Temporal Dynamics for Multivariate Time Series Forecasting with Missing Values

2020 ◽  
Vol 34 (04) ◽  
pp. 5956-5963
Author(s):  
Xianfeng Tang ◽  
Huaxiu Yao ◽  
Yiwei Sun ◽  
Charu Aggarwal ◽  
Prasenjit Mitra ◽  
...  
Keyword(s):  
Time Series ◽  
Real World ◽  
Missing Values ◽  
Temporal Dynamics ◽  
Multivariate Time Series ◽  
Temporal Distribution ◽  
Joint Modeling ◽  
Adversarial Training ◽  
Global Patterns ◽  
Real World Datasets

Multivariate time series (MTS) forecasting is widely used in various domains, such as meteorology and traffic. Due to limitations on data collection, transmission, and storage, real-world MTS data usually contains missing values, making it infeasible to apply existing MTS forecasting models such as linear regression and recurrent neural networks. Though many efforts have been devoted to this problem, most of them solely rely on local dependencies for imputing missing values, which ignores global temporal dynamics. Local dependencies/patterns would become less useful when the missing ratio is high, or the data have consecutive missing values; while exploring global patterns can alleviate such problem. Thus, jointly modeling local and global temporal dynamics is very promising for MTS forecasting with missing values. However, work in this direction is rather limited. Therefore, we study a novel problem of MTS forecasting with missing values by jointly exploring local and global temporal dynamics. We propose a new framework øurs, which leverages memory network to explore global patterns given estimations from local perspectives. We further introduce adversarial training to enhance the modeling of global temporal distribution. Experimental results on real-world datasets show the effectiveness of øurs for MTS forecasting with missing values and its robustness under various missing ratios.

scholarly journals DATA-GRU: Dual-Attention Time-Aware Gated Recurrent Unit for Irregular Multivariate Time Series

2020 ◽  
Vol 34 (01) ◽  
pp. 930-937
Author(s):  
Qingxiong Tan ◽  
Mang Ye ◽  
Baoyao Yang ◽  
Siqi Liu ◽  
Andy Jinhua Ma ◽  
...  
Keyword(s):  
Time Series ◽  
Missing Values ◽  
Multivariate Time Series ◽  
Medical Knowledge ◽  
Attention Mechanism ◽  
Physiological Variables ◽  
Real World Datasets ◽  
Clinical Records ◽  
Gated Recurrent Unit ◽  
Time Aware

Due to the discrepancy of diseases and symptoms, patients usually visit hospitals irregularly and different physiological variables are examined at each visit, producing large amounts of irregular multivariate time series (IMTS) data with missing values and varying intervals. Existing methods process IMTS into regular data so that standard machine learning models can be employed. However, time intervals are usually determined by the status of patients, while missing values are caused by changes in symptoms. Therefore, we propose a novel end-to-end Dual-Attention Time-Aware Gated Recurrent Unit (DATA-GRU) for IMTS to predict the mortality risk of patients. In particular, DATA-GRU is able to: 1) preserve the informative varying intervals by introducing a time-aware structure to directly adjust the influence of the previous status in coordination with the elapsed time, and 2) tackle missing values by proposing a novel dual-attention structure to jointly consider data-quality and medical-knowledge. A novel unreliability-aware attention mechanism is designed to handle the diversity in the reliability of different data, while a new symptom-aware attention mechanism is proposed to extract medical reasons from original clinical records. Extensive experimental results on two real-world datasets demonstrate that DATA-GRU can significantly outperform state-of-the-art methods and provide meaningful clinical interpretation.

scholarly journals A Novel LSTM for Multivariate Time Series with Massive Missingness

Sensors ◽  
2020 ◽  
Vol 20 (10) ◽  
pp. 2832
Author(s):  
Nazanin Fouladgar ◽  
Kary Främling
Keyword(s):  
Time Series ◽  
Short Term Memory ◽  
Time Series Prediction ◽  
Multivariate Time Series ◽  
Streaming Data ◽  
Short Term ◽  
Time Intervals ◽  
Long Short Term Memory ◽  
Real World Datasets ◽  
Novel Model

Multivariate time series with missing data is ubiquitous when the streaming data is collected by sensors or any other recording instruments. For instance, the outdoor sensors gathering different meteorological variables may encounter low material sensitivity to specific situations, leading to incomplete information gathering. This is problematic in time series prediction with massive missingness and different missing rate of variables. Contribution addressing this problem on the regression task of meteorological datasets by employing Long Short-Term Memory (LSTM), capable of controlling the information flow with its memory unit, is still missing. In this paper, we propose a novel model called forward and backward variable-sensitive LSTM (FBVS-LSTM) consisting of two decay mechanisms and some informative data. The model inputs are mainly the missing indicator, time intervals of missingness in both forward and backward direction and missing rate of each variable. We employ this information to address the so-called missing not at random (MNAR) mechanism. Separately learning the features of each parameter, the model becomes adapted to deal with massive missingness. We conduct our experiment on three real-world datasets for the air pollution forecasting. The results demonstrate that our model performed well along with other LSTM-derivation models in terms of prediction accuracy.

scholarly journals A Novel Deep Learning Approach for Anomaly Detection of Time Series Data

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Zhiwei Ji ◽  
Jiaheng Gong ◽  
Jiarui Feng
Keyword(s):  
Time Series ◽  
Anomaly Detection ◽  
Time Series Data ◽  
Short Term Memory ◽  
Training Data ◽  
Series Data ◽  
Real Time System ◽  
Statistical Strategy ◽  
Anomaly Classification ◽  
Real World Datasets

Anomalies in time series, also called “discord,” are the abnormal subsequences. The occurrence of anomalies in time series may indicate that some faults or disease will occur soon. Therefore, development of novel computational approaches for anomaly detection (discord search) in time series is of great significance for state monitoring and early warning of real-time system. Previous studies show that many algorithms were successfully developed and were used for anomaly classification, e.g., health monitoring, traffic detection, and intrusion detection. However, the anomaly detection of time series was not well studied. In this paper, we proposed a long short-term memory- (LSTM-) based anomaly detection method (LSTMAD) for discord search from univariate time series data. LSTMAD learns the structural features from normal (nonanomalous) training data and then performs anomaly detection via a statistical strategy based on the prediction error for observed data. In our experimental evaluation using public ECG datasets and real-world datasets, LSTMAD detects anomalies more accurately than other existing approaches in comparison.

scholarly journals SYNERGISTIC USE OF SENTINEL-1 AND SENTINEL-2 TIME SERIES FOR POPLAR PLANTATIONS MONITORING AT LARGE SCALE

2020 ◽  
Vol XLIII-B3-2020 ◽  
pp. 1457-1461
Author(s):  
Y. Hamrouni ◽  
É. Paillassa ◽  
V. Chéret ◽  
C. Monteil ◽  
D. Sheeren
Keyword(s):  
Time Series ◽  
Active Learning ◽  
Supervised Classification ◽  
Large Scale ◽  
Training Data ◽  
Passive Learning ◽  
Training Samples ◽  
Poplar Plantations ◽  
Annual Means ◽  
Sentinel 2

Abstract. The current context of availability of Earth Observation satellite data at high spatial and temporal resolutions makes it possible to map large areas. Although supervised classification is the most widely adopted approach, its performance is highly dependent on the availability and the quality of training data. However, gathering samples from field surveys or through photo interpretation is often expensive and time-consuming especially when the area to be classified is large. In this paper we propose the use of an active learning-based technique to address this issue by reducing the labelling effort required for supervised classification while increasing the generalisation capabilities of the classifier across space. Experiments were conducted to identify poplar plantations in three different sites in France using Sentinel-2 time series. In order to characterise the age of the identified poplar stands, temporal means of Sentinel-1 backscatter coefficients were computed. The results are promising and show the good capacities of the active learning-based approach to achieve similar performance (Poplar F-score ≥ 90%) to traditional passive learning (i.e. with random selection of samples) with up to 50% fewer training samples. Sentinel-1 annual means have demonstrated their potential to differentiate two stand ages with an overall accuracy of 83% regardless of the cultivar considered.

scholarly journals Time Series GIS Map Dataset of Demolished Buildings in Mashiki Town after the 2016 Kumamoto, Japan Earthquake

2019 ◽  
Vol 11 (19) ◽  
pp. 2190
Author(s):  
Kushiyama ◽  
Matsuoka
Keyword(s):  
Time Series ◽  
Waste Disposal ◽  
Large Scale ◽  
Training Data ◽  
Training Dataset ◽  
Building Structures ◽  
Visual Interpretation ◽  
Imbalanced Classification ◽  
Map Data ◽  
F Measure

After a large-scale disaster, many damaged buildings are demolished and treated as disaster waste. Though the weight of disaster waste was estimated two months after the 2016 earthquake in Kumamoto, Japan, the estimated weight was significantly different from the result when the disaster waste disposal was completed in March 2018. The amount of disaster waste generated is able to be estimated by an equation by multiplying the total number of severely damaged and partially damaged buildings by the coefficient of generated weight per building. We suppose that the amount of disaster waste would be affected by the conditions of demolished buildings, namely, the areas and typologies of building structures, but this has not yet been clarified. Therefore, in this study, we aimed to use geographic information system (GIS) map data to create a time series GIS map dataset with labels of demolished and remaining buildings in Mashiki town for the two-year period prior to the completion of the disaster waste disposal. We used OpenStreetMap (OSM) data as the base data and time series SPOT images observed in the two years following the Kumamoto earthquake to label all demolished and remaining buildings in the GIS map dataset. To effectively label the approximately 16,000 buildings in Mashiki town, we calculated an indicator that shows the possibility of the buildings to be classified as the remaining and demolished buildings from a change of brightness in SPOT images. We classified 5701 demolished buildings from 16,106 buildings, as of March 2018, by visual interpretation of the SPOT and Pleiades images with reference to this indicator. We verified that the number of demolished buildings was almost the same as the number reported by Mashiki municipality. Moreover, we assessed the accuracy of our proposed method: The F-measure was higher than 0.9 using the training dataset, which was verified by a field survey and visual interpretation, and included the labels of the 55 demolished and 55 remaining buildings. We also assessed the accuracy of the proposed method by applying it to all the labels in the OSM dataset, but the F-measure was 0.579. If we applied test data including balanced labels of the other 100 demolished and 100 remaining buildings, which were other than the training data, the F-measure was 0.790 calculated from the SPOT image of 25 March 2018. Our proposed method performed better for the balanced classification but not for imbalanced classification. We studied the examples of image characteristics of correct and incorrect estimation by thresholding the indicator.

scholarly journals Learning Meta Model for Zero- and Few-Shot Face Anti-Spoofing

2020 ◽  
Vol 34 (07) ◽  
pp. 11916-11923 ◽  
Author(s):  
Yunxiao Qin ◽  
Chenxu Zhao ◽  
Xiangyu Zhu ◽  
Zezheng Wang ◽  
Zitong Yu ◽  
...  
Keyword(s):  
Face Recognition ◽  
Supervised Learning ◽  
Large Scale ◽  
Training Data ◽  
Learning Problem ◽  
Meta Model ◽  
Meta Learning ◽  
Recognition Systems

Face anti-spoofing is crucial to the security of face recognition systems. Most previous methods formulate face anti-spoofing as a supervised learning problem to detect various predefined presentation attacks, which need large scale training data to cover as many attacks as possible. However, the trained model is easy to overfit several common attacks and is still vulnerable to unseen attacks. To overcome this challenge, the detector should: 1) learn discriminative features that can generalize to unseen spoofing types from predefined presentation attacks; 2) quickly adapt to new spoofing types by learning from both the predefined attacks and a few examples of the new spoofing types. Therefore, we define face anti-spoofing as a zero- and few-shot learning problem. In this paper, we propose a novel Adaptive Inner-update Meta Face Anti-Spoofing (AIM-FAS) method to tackle this problem through meta-learning. Specifically, AIM-FAS trains a meta-learner focusing on the task of detecting unseen spoofing types by learning from predefined living and spoofing faces and a few examples of new attacks. To assess the proposed approach, we propose several benchmarks for zero- and few-shot FAS. Experiments show its superior performances on the presented benchmarks to existing methods in existing zero-shot FAS protocols.

scholarly journals Improving Multi-Agent Generative Adversarial Nets with Variational Latent Representation

Entropy ◽  
2020 ◽  
Vol 22 (9) ◽  
pp. 1055
Author(s):  
Huan Zhao ◽  
Tingting Li ◽  
Yufeng Xiao ◽  
Yu Wang
Keyword(s):  
Large Scale ◽  
Random Noise ◽  
Qualitative Evaluation ◽  
Training Data ◽  
Generative Adversarial Networks ◽  
Generative Adversarial Network ◽  
Proposed Model ◽  
Multi Agent ◽  
Latent Representations ◽  
Real World Datasets

Generative adversarial networks (GANs), which are a promising type of deep generative network, have recently drawn considerable attention and made impressive progress. However, GAN models suffer from the well-known problem of mode collapse. This study focuses on this challenge and introduces a new model design, called the encoded multi-agent generative adversarial network (E-MGAN), which tackles the mode collapse problem by introducing the variational latent representations learned from a variable auto-encoder (VAE) to a multi-agent GAN. The variational latent representations are extracted from training data to replace the random noise input of the general multi-agent GANs. The generator in E-MGAN employs multiple generators and is penalized by a classifier. This integration guarantees that the proposed model not only enhances the quality of generated samples but also improves the diversity of generated samples to avoid the mode collapse problem. Moreover, extensive experiments are conducted on both a synthetic dataset and two large-scale real-world datasets. The generated samples are visualized for qualitative evaluation. The inception score (IS) and Fréchet inception distance (FID) are adopted to measure the performance of the model for quantitative assessment. The results confirmed that the proposed model achieves outstanding performances compared to other state-of-the-art GAN variants.

Sign in / Sign up

Export Citation Format

Share Document