MBGAN: An improved generative adversarial network with multi-head self-attention and bidirectional RNN for time series imputation

Ni, Qingjian; Cao, Xuehan

doi:10.1016/j.engappai.2022.105232

Cited by 28 publications

(4 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MBGAN is a time series imputation model proposed by Ni etal in 2020 [11]. The main framework of this model is still GAN.…”

Section: Mbganmentioning

confidence: 99%

“…The essence of these methods is to use the data near the missing point as the feature, predict the data of the missing point, and mine similar change models from a large number of historical data, so as to carry out more accurate data filling. E 2 GAN [8], BIGAN [9,10] and MBGAN [11] are typical models for imputation and filling of time series data using GAN, and we will further compare and analyze in the following.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Research on Time Series Imputation Based on Generative Adversarial Network

Xiao,

2023

Advances in Transdisciplinary Engineering

View full text Add to dashboard Cite

Time series data is collected in chronological order to represent how the collected data changes over time. This type of data is susceptible to interference from external factors that ultimately make the data missing. Missing data will cause the lack of some historical information, and it is not conducive to the development of downstream tasks. In recent years, Generative Adversarial Networks (GAN) GAN is widely used in image processing tasks and has achieved good results, which can also be applied to time series generation and interpolation tasks. We analyze the types of time series data missing, the structure of general GAN, and compare three models such as E2GAN, BIGAN and MBGAN, focusing on the comparison of the model composition of generator and discriminator, so as to provide ideas for the subsequent optimization application of GAN in time series data imputation.

show abstract

“…MBGAN is a time series imputation model proposed by Ni etal in 2020 [11]. The main framework of this model is still GAN.…”

Section: Mbganmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Research on Time Series Imputation Based on Generative Adversarial Network

Xiao,

2023

Advances in Transdisciplinary Engineering

View full text Add to dashboard Cite

show abstract

“…In recent years, deep-generative models, such as GANs, have emerged as invaluable tools to generate high-quality samples, beneficial for imputation and classification tasks [6][7][8]. However, many GAN-based methods tend to disregard the static features of the population present in the…”

Section: Introductionmentioning

confidence: 99%

Joint Representation Learning with Generative Adversarial Imputation Network for Improved Classification of Longitudinal Data

Pingi,

Zhang,

Bashar

et al. 2023

Data Sci. Eng.

View full text Add to dashboard Cite

Generative adversarial networks (GANs) have demonstrated their effectiveness in generating temporal data to fill in missing values, enhancing the classification performance of time series data. Longitudinal datasets encompass multivariate time series data with additional static features that contribute to sample variability over time. These datasets often encounter missing values due to factors such as irregular sampling. However, existing GAN-based imputation methods that address this type of data missingness often overlook the impact of static features on temporal observations and classification outcomes. This paper presents a novel method, fusion-aided imputer-classifier GAN (FaIC-GAN), tailored for longitudinal data classification. FaIC-GAN simultaneously leverages partially observed temporal data and static features to enhance imputation and classification learning. We present four multimodal fusion strategies that effectively extract correlated information from both static and temporal modalities. Our extensive experiments reveal that FaIC-GAN successfully exploits partially observed temporal data and static features, resulting in improved classification accuracy compared to unimodal models. Our post-additive and attention-based multimodal fusion approaches within the FaIC-GAN model consistently rank among the top three methods for classification.

show abstract

“…Generative models [21,29] have shown outstanding performance in the image domain, and some efforts have been made to apply them to the task of TSDI to generate more realistic imputed values [2]. A two-stage GAN [16] imputation method was proposed by combining GAN with a new RNN unit, namely GRUI, which learns the distribution of time series data to optimize the generator's input vector.…”

Section: Introductionmentioning

confidence: 99%

Time-Series Data Imputation via Realistic Masking-Guided Tri-Attention Bi-GRU

Zhang,

Zeng

et al. 2023

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

Time series data with missing values are ubiquitous in real applications due to various unforeseen faults during data generation, storage, and transmission. Time-Series Data Imputation (TSDI) is thus crucial to many temporal data analysis tasks. However, existing works usually consider only one of the following two issues: (1) intra-feature temporal dependency, and (2) inter-feature correlation, leading to the overlook of complex coupling information in imputation. To achieve more accurate TDSI, we design a novel imputation model called TABiG, which delicately preserves the short-term, long-term, and inter-feature dependencies by attention mechanisms in a delay error-reduced bi-directional architecture. That is, it leverages GRU to model short-term temporal dependencies and adopts self-attention mechanisms hierarchically to capture long-term temporal dependencies and inter-feature correlations. The multiple self-attention mechanisms are nested in a bi-directional structure to alleviate the problem of delay errors in RNN-like structures. To facilitate model training with higher generalization, a masking strategy that mimics various extreme real missing situations beyond the simple random ones has been adopted for generating self-supervised learning tasks. Comprehensive experiments demonstrate that TABiG significantly outperforms most state-of-the-art imputation counterparts. Complementary results and source code can be accessed at https://github.com/Zhang2112105189/TABiG

show abstract

MBGAN: An improved generative adversarial network with multi-head self-attention and bidirectional RNN for time series imputation

Cited by 28 publications

References 29 publications

Research on Time Series Imputation Based on Generative Adversarial Network

Research on Time Series Imputation Based on Generative Adversarial Network

Joint Representation Learning with Generative Adversarial Imputation Network for Improved Classification of Longitudinal Data

Time-Series Data Imputation via Realistic Masking-Guided Tri-Attention Bi-GRU

Contact Info

Product

Resources

About