A Flexible Approach for Human Activity Recognition Based on Broad Learning System

Lin, Zhidi; Chen, Haipeng; Yang, Qi; Hong, Xuemin

doi:10.1145/3318299.3318318

Cited by 5 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the online training stage, only the mutual information network is fine-tuned with the incoming new data stream. We note that if the new incoming data is insufficient, an oversampling technique as introduced in [43] could be used to further enhance the training performance.…”

Section: B Unsupervised Feature Extraction Based On Dimmentioning

confidence: 99%

Baseline Model Training in Sensor-Based Human Activity Recognition: An Incremental Learning Approach

Xiao

Chen

et al. 2021

IEEE Access

Self Cite

View full text Add to dashboard Cite

Human activity recognition (HAR) based on wearable sensors has attracted significant research attention in recent years due to its advantages in availability, accuracy, and privacy-friendliness. HAR baseline model is essentially a general-purpose classifier trained to recognized multiple activity patterns of most user types. It provides the input for subsequent steps of model personalization. Training a good baseline model is of fundamental importance because it has significant impacts on the ultimate HAR accuracy. In practice, baseline model training in HAR is a non-trivial problem that faces two challenges: insufficient training data and biased training data. This paper proposes a novel baseline model training scheme to tackle the two challenges using Deep InfoMax (DIM)-based unsupervised feature extraction and Broad Learning System (BLS)-based incremental learning, respectively. Experimental results demonstrate that the proposed scheme outperform conventional methods in terms of overall accuracy, computational efficiency, and the ability to adapt to dynamic scenarios with changing data characteristics. INDEX TERMS Human activity recognition, baseline model, broad learning system, DIM, incremental learning.

show abstract

Section: B Unsupervised Feature Extraction Based On Dimmentioning

confidence: 99%

Baseline Model Training in Sensor-Based Human Activity Recognition: An Incremental Learning Approach

Xiao

Chen

et al. 2021

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…With the continuous improvement of BLS, some researchers have used BLS to achieve impressive performance in image processing and other fields. Lin et al in [35] proposed a flexible approach for human activity recognition based on BLS, which improves the model training speed and prediction accuracy by introducing BLS. Kong et al in [36] proposed an image classification method based on semi-supervised BLS.…”

Section: Bls Development Historymentioning

confidence: 99%

Privacy-Preserving Cloud-Aided Broad Learning System

Liu¹,

Zhang²,

Guo³

et al. 2021

Preprint

View full text Add to dashboard Cite

With the rapid development of artificial intelligence and the advent of the 5G era, deep learning has received extensive attention from researchers. Broad Learning System (BLS) is a new deep learning model proposed recently, which shows its effectiveness in many fields, such as image recognition and fault detection. However, the training process still requires vast computations, and therefore cannot be accomplished by some resource-constrained devices. To solve this problem, the resource-constrained device can outsource the BLS algorithm to cloud servers. Nevertheless, some security challenges also follow with the use of cloud computing, including the privacy of the data and the correctness of returned results. In this paper, we propose a secure, efficient, and verifiable outsourcing algorithm for BLS. This algorithm not only improves the efficiency of the algorithm on the client but also ensures that the clients sensitive information is not leaked to the cloud server. In addition, in our algorithm, the client can verify the correctness of returned results with a probability of almost 1. Finally, we analyze the security and efficiency of our algorithm in theory and prove our algorithms feasibility through experiments.

show abstract

“…Zhang et al [36] proposed a BLS model for Facial Expression Recognition (FER). Lin et al [37] proposed an incremental BLS model for human activity recognition. Xu et al [38] proposed a deep-and-wide network to predict lncRNA subcellular locations.…”

Section: Introductionmentioning

confidence: 99%

State-of-Health Prediction for Reaction Wheel of On-Orbit Satellite Based on Fourier Broad Learning System

Zhang

Wang

et al. 2021

IEEE Access

View full text Add to dashboard Cite

The state-of-health (SoH) of the reaction wheel (RW) is a key indicator of the satellite's ability to function properly and complete its mission successfully. Quick and accurate prediction of the RW's SoH is an important guarantee of autonomous satellite mission execution and health management. In view of the pseudoperiodicity and strong noise characteristics of satellite on-orbit telemetry data, this paper aims to design a prediction framework meeting the satellite state on-orbit prediction requirements, which mainly includes data reconstruction, similar data screening, self-revised prediction and the fuzzy expression of results. First, a new method for RW health characterization is proposed for actual telemetry data from satellites, and a completely new set of historical samples is constructed. Second, a fusion deviation measure is proposed by selecting samples from the historical sample set that are similar to the sample to be predicted. Third, a Fourier Broad Learning System (FBLS) is proposed to improve the depiction accuracy of a standard Broad Learning System (BLS) in learning satellite telemetry data with less computational and time resources. Finally, a fuzzy expression is proposed to obtain the prediction results. The proposed method is validated with real satellite on-orbit telemetry data. The case study results show that the SoH of the RW can be predicted quickly and accurately, which demonstrates the promising performance of our method.

show abstract

A Flexible Approach for Human Activity Recognition Based on Broad Learning System

Cited by 5 publications

References 17 publications

Baseline Model Training in Sensor-Based Human Activity Recognition: An Incremental Learning Approach

Baseline Model Training in Sensor-Based Human Activity Recognition: An Incremental Learning Approach

Privacy-Preserving Cloud-Aided Broad Learning System

State-of-Health Prediction for Reaction Wheel of On-Orbit Satellite Based on Fourier Broad Learning System

Contact Info

Product

Resources

About