scholarly journals A new approach to extended‐range multi‐model forecasting: sequential learning algorithms

2021 ◽  
Author(s):  
Paula L.M. Gonzalez ◽  
David J. Brayshaw ◽  
Florian Ziel
Keyword(s):  
Learning Algorithms ◽  
Sequential Learning ◽  
New Approach ◽  
Extended Range

Developing a new approach for design support of subsurface constructed wetland using machine learning algorithms

2022 ◽  
Vol 301 ◽  
pp. 113868
Author(s):  
Xuan Cuong Nguyen ◽  
Thi Thanh Huyen Nguyen ◽  
Quyet V. Le ◽  
Phuoc Cuong Le ◽  
Arun Lal Srivastav ◽  
...  
Keyword(s):  
Machine Learning ◽  
Constructed Wetland ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Design Support ◽  
New Approach

Collaborative Calibrated Peer Assessment in Massive Open Online Courses

2020 ◽  
pp. 1408-1434
Author(s):  
Asma Boudria ◽  
Yacine Lafifi ◽  
Yamina Bordjiba
Keyword(s):  
Open Access ◽  
Collaborative Learning ◽  
Online Courses ◽  
Peer Assessment ◽  
Learning Algorithms ◽  
Massive Open Online Courses ◽  
Major Difficulty ◽  
Massive Open Online ◽  
New Approach

The free nature and open access courses in the Massive Open Online Courses (MOOC) allow the facilities of disseminating information for a large number of participants. However, the “massive” propriety can generate many pedagogical problems, such as the assessment of learners, which is considered as the major difficulty facing in the MOOC. In fact, the immense number of learners who exceeded in some MOOC the hundreds of thousands make the instructors' evaluation of students' production quite impossible. In this work, the authors present a new approach for assessing the learners' production in MOOC. This approach combines the peer assessment with the collaborative learning and the calibrated method. It aims at increasing the degree of trust in peer-assessment. For evaluating the proposed approach, the authors implemented a MOOC dedicated for learning algorithms. In addition, an experiment was conducted during two months for knowing the effects of the proposed approach. The obtained results are presented in this paper. They are judged as very interesting and encouraging.

scholarly journals Computationally efficient sequential learning algorithms for direct link resource-allocating networks

2005 ◽  
Vol 69 (1-3) ◽  
pp. 142-157 ◽  
Author(s):  
Vijanth S. Asirvadam ◽  
Seán F. McLoone ◽  
George W. Irwin
Keyword(s):  
Learning Algorithms ◽  
Sequential Learning ◽  
Direct Link ◽  
Computationally Efficient

Local Learning Algorithms for Sequential Tasks in Neural Networks

1998 ◽  
Vol 2 (6) ◽  
pp. 221-227 ◽  
Author(s):  
Anthony Robins ◽  
◽  
Marcus Frean ◽  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Learning Algorithms ◽  
Learning Task ◽  
Sequential Learning ◽  
Local Learning ◽  
Neural Network Learning ◽  
Network Learning ◽  
Sequential Task ◽  
Global And Local

In this paper, we explore the concept of sequential learning and the efficacy of global and local neural network learning algorithms on a sequential learning task. Pseudorehearsal, a method developed by Robins19) to solve the catastrophic forgetting problem which arises from the excessive plasticity of neural networks, is significantly more effective than other local learning algorithms for the sequential task. We further consider the concept of local learning and suggest that pseudorehearsal is so effective because it works directly at the level of the learned function, and not indirectly on the representation of the function within the network. We also briefly explore the effect of local learning on generalization within the task.

scholarly journals Decadal Climate Predictions Using Sequential Learning Algorithms

2016 ◽  
Vol 29 (10) ◽  
pp. 3787-3809 ◽  
Author(s):  
Ehud Strobach ◽  
Golan Bel
Keyword(s):  
Climate Models ◽  
Statistical Significance ◽  
Learning Algorithms ◽  
Meridional Wind ◽  
Sequential Learning ◽  
Forecast Errors ◽  
Physical Processes ◽  
Additional Information ◽  
Decadal Climate ◽  
Better Than

Abstract Ensembles of climate models are commonly used to improve decadal climate predictions and assess the uncertainties associated with them. Weighting the models according to their performances holds the promise of further improving their predictions. Here, an ensemble of decadal climate predictions is used to demonstrate the ability of sequential learning algorithms (SLAs) to reduce the forecast errors and reduce the uncertainties. Three different SLAs are considered, and their performances are compared with those of an equally weighted ensemble, a linear regression, and the climatology. Predictions of four different variables—the surface temperature, the zonal and meridional wind, and pressure—are considered. The spatial distributions of the performances are presented, and the statistical significance of the improvements achieved by the SLAs is tested. The reliability of the SLAs is also tested, and the advantages and limitations of the different measures of the performance are discussed. It was found that the best performances of the SLAs are achieved when the learning period is comparable to the prediction period. The spatial distribution of the SLAs performance showed that they are skillful and better than the other forecasting methods over large continuous regions. This finding suggests that, despite the fact that each of the ensemble models is not skillful, they were able to capture some physical processes that resulted in deviations from the climatology and that the SLAs enabled the extraction of this additional information.

Sign in / Sign up

Export Citation Format

Share Document