Hint: harnessing the wisdom of crowds for handling multi-phase tasks

Fang, Yili; Chen, Pengpeng; Han, Tao

doi:10.1007/s00521-021-06825-7

Cited by 7 publications

(6 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Defining BCI illiteracy has been challenging because different researchers use different criteria to distinguish between good and bad subjects. Early work stated that 70% accuracy was effective for binary classification [ 58 , 59 ]. In Table 2 , the nontransfer learning algorithm CSP-LDA reaches or approaches a benchmark accuracy of 70% for subjects al, aw, and ay.…”

Section: Resultsmentioning

confidence: 99%

Supervised and Semisupervised Manifold Embedded Knowledge Transfer in Motor Imagery-Based BCI

Yin

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

A long calibration procedure limits the use in practice for a motor imagery (MI)-based brain-computer interface (BCI) system. To tackle this problem, we consider supervised and semisupervised transfer learning. However, it is a challenge for them to cope with high intersession/subject variability in the MI electroencephalographic (EEG) signals. Based on the framework of unsupervised manifold embedded knowledge transfer (MEKT), we propose a supervised MEKT algorithm (sMEKT) and a semisupervised MEKT algorithm (ssMEKT), respectively. sMEKT only has limited labelled samples from a target subject and abundant labelled samples from multiple source subjects. Compared to sMEKT, ssMEKT adds comparably more unlabelled samples from the target subject. After performing Riemannian alignment (RA) and tangent space mapping (TSM), both sMEKT and ssMEKT execute domain adaptation to shorten the differences among subjects. During domain adaptation, to make use of the available samples, two algorithms preserve the source domain discriminability, and ssMEKT preserves the geometric structure embedded in the labelled and unlabelled target domains. Moreover, to obtain a subject-specific classifier, sMEKT minimizes the joint probability distribution shift between the labelled target and source domains, whereas ssMEKT performs the joint probability distribution shift minimization between the unlabelled target domain and all labelled domains. Experimental results on two publicly available MI datasets demonstrate that our algorithms outperform the six competing algorithms, where the sizes of labelled and unlabelled target domains are variable. Especially for the target subjects with 10 labelled samples and 270/190 unlabelled samples, ssMEKT shows 5.27% and 2.69% increase in average accuracy on the two abovementioned datasets compared to the previous best semisupervised transfer learning algorithm (RA-regularized common spatial patterns-weighted adaptation regularization, RA-RCSP-wAR), respectively. Therefore, our algorithms can effectively reduce the need of labelled samples for the target subject, which is of importance for the MI-based BCI application.

show abstract

Section: Resultsmentioning

confidence: 99%

Supervised and Semisupervised Manifold Embedded Knowledge Transfer in Motor Imagery-Based BCI

Yin

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…In a ring signature, the number of public keys cannot be less than three, so n ≥ 3, n ∈ N * . A 2 > A 3 can be obtained by using (10) and (11). As n increases, the gap between the two ring signature schemes increases in terms of storage space.…”

Section: Evaluation Methodsmentioning

confidence: 99%

“…Currently, the data for these applications are extracted using big data and aggregation analysis. These techniques may be at risk of privacy data leakage [10]. It is possible to store a wide variety of data, including financial information, identity, assets of an individual, etc., in a blockchain [11].…”

Section: Introductionmentioning

confidence: 99%

Representative Ring Signature Algorithm Based on Smart Contract

Yi²,

Zhao³

et al. 2022

Sensors

View full text Add to dashboard Cite

Traditional ring signature algorithms suffer from large signature data capacity and low speed of signature and verification during collective signing. In this work, we propose a representative ring signature algorithm based on smart contracts. By collecting the opinions of the signatory based on multiparty secure computation, the proposed technique protects the privacy of the signatory during the data interaction process in the consortium chain. Moreover, the proposed method uses smart contracts to organize the signature process and formulate a signature strategy of “one encryption per signature” to prevent signature forgery. It uses the Hyperledger Fabric framework as the signature test platform of the consortium chain to perform the experiments. We compare the results of the proposed method with the ECC ring signature scheme. The experimental results show that in the worst case, the signature volume of the proposed method decreases by more than two times, and the signature speed and verification speed increase by more than three times. Therefore, in the collective signature scenario of transaction verification in the consortium chain, the proposed method is verified to be innovative and practical.

show abstract

“…Educators should receive education first, and infect students with their noble moral character, excellent professional skills, and selfless dedication to their work. Being a role model and teaching by example can serve as a good example [10][11][12].…”

Section: Reflections On the Integration Of Ideological And Political ...mentioning

confidence: 99%

Exploration of the Curriculum Reform of "Logistics System Simulation" Based on Chinese-foreign Cooperation in Running Schools

2023

IJNDE

View full text Add to dashboard Cite

Faced with the requirements of curriculum reform in Chinese-foreign cooperation in running schools, this article explores the reform of the compulsory courses "Logistics System Simulation" for logistics engineering majors. Based on the OBE (Outcome based education) education philosophy, with students as the center, specific goals are proposed for course learning in terms of knowledge, application, integration, emotion, value and learning. Then, teaching contents are designed and teaching methods are studied. Finally, course assessment is explored.

show abstract

Hint: harnessing the wisdom of crowds for handling multi-phase tasks

Cited by 7 publications

References 63 publications

Supervised and Semisupervised Manifold Embedded Knowledge Transfer in Motor Imagery-Based BCI

Supervised and Semisupervised Manifold Embedded Knowledge Transfer in Motor Imagery-Based BCI

Representative Ring Signature Algorithm Based on Smart Contract

Exploration of the Curriculum Reform of "Logistics System Simulation" Based on Chinese-foreign Cooperation in Running Schools

Contact Info

Product

Resources

About