Recently,the convolutional neural network(CNN)has been dom-inant in studies on interpreting remote sensing images(RSI).However,it appears that training optimization strategies have received less attention in relevant ...Recently,the convolutional neural network(CNN)has been dom-inant in studies on interpreting remote sensing images(RSI).However,it appears that training optimization strategies have received less attention in relevant research.To evaluate this problem,the author proposes a novel algo-rithm named the Fast Training CNN(FST-CNN).To verify the algorithm’s effectiveness,twenty methods,including six classic models and thirty archi-tectures from previous studies,are included in a performance comparison.The overall accuracy(OA)trained by the FST-CNN algorithm on the same model architecture and dataset is treated as an evaluation baseline.Results show that there is a maximal OA gap of 8.35%between the FST-CNN and those methods in the literature,which means a 10%margin in performance.Meanwhile,all those complex roadmaps,e.g.,deep feature fusion,model combination,model ensembles,and human feature engineering,are not as effective as expected.It reveals that there was systemic suboptimal perfor-mance in the previous studies.Most of the CNN-based methods proposed in the previous studies show a consistent mistake,which has made the model’s accuracy lower than its potential value.The most important reasons seem to be the inappropriate training strategy and the shift in data distribution introduced by data augmentation(DA).As a result,most of the performance evaluation was conducted based on an inaccurate,suboptimal,and unfair result.It has made most of the previous research findings questionable to some extent.However,all these confusing results also exactly demonstrate the effectiveness of FST-CNN.This novel algorithm is model-agnostic and can be employed on any image classification model to potentially boost performance.In addition,the results also show that a standardized training strategy is indeed very meaningful for the research tasks of the RSI-SC.展开更多
Enlightened by Mal’cev theorem in universal algebra, a new criterion for consistency argument in λ-calculus has been introduced. It is equivalent to Jacopini and Baeten-Boerboom’ s, but more convenient to use. Base...Enlightened by Mal’cev theorem in universal algebra, a new criterion for consistency argument in λ-calculus has been introduced. It is equivalent to Jacopini and Baeten-Boerboom’ s, but more convenient to use. Based on the new criterion, one uses an enhanced technique to show a few results which provides a deeper insight in the classification problem of λ-terms with no normal forms.展开更多
The rough sets and Boolean reasoning based discretization approach (RSBRA) is no t suitable for feature selection for machine learning algorithms such as neural network or SVM because the information loss due to discr...The rough sets and Boolean reasoning based discretization approach (RSBRA) is no t suitable for feature selection for machine learning algorithms such as neural network or SVM because the information loss due to discretization is large. A mo dified RSBRA for feature selection was proposed and evaluated with SVM classifie rs. In the presented algorithm, the level of consistency, coined from the rough sets theory, is introduced to substitute the stop criterion of circulation of th e RSBRA, which maintains the fidelity of the training set after discretization. The experimental results show the modified algorithm has better predictive accur acy and less training time than the original RSBRA.展开更多
基金Hunan University of Arts and Science provided doctoral research funding for this study (grant number 16BSQD23)Fund of Geography Subject ([2022]351)also provided funding.
文摘Recently,the convolutional neural network(CNN)has been dom-inant in studies on interpreting remote sensing images(RSI).However,it appears that training optimization strategies have received less attention in relevant research.To evaluate this problem,the author proposes a novel algo-rithm named the Fast Training CNN(FST-CNN).To verify the algorithm’s effectiveness,twenty methods,including six classic models and thirty archi-tectures from previous studies,are included in a performance comparison.The overall accuracy(OA)trained by the FST-CNN algorithm on the same model architecture and dataset is treated as an evaluation baseline.Results show that there is a maximal OA gap of 8.35%between the FST-CNN and those methods in the literature,which means a 10%margin in performance.Meanwhile,all those complex roadmaps,e.g.,deep feature fusion,model combination,model ensembles,and human feature engineering,are not as effective as expected.It reveals that there was systemic suboptimal perfor-mance in the previous studies.Most of the CNN-based methods proposed in the previous studies show a consistent mistake,which has made the model’s accuracy lower than its potential value.The most important reasons seem to be the inappropriate training strategy and the shift in data distribution introduced by data augmentation(DA).As a result,most of the performance evaluation was conducted based on an inaccurate,suboptimal,and unfair result.It has made most of the previous research findings questionable to some extent.However,all these confusing results also exactly demonstrate the effectiveness of FST-CNN.This novel algorithm is model-agnostic and can be employed on any image classification model to potentially boost performance.In addition,the results also show that a standardized training strategy is indeed very meaningful for the research tasks of the RSI-SC.
基金Project supported by the National 973 project of China: Mechanization of mathematics and Automata platform.
文摘Enlightened by Mal’cev theorem in universal algebra, a new criterion for consistency argument in λ-calculus has been introduced. It is equivalent to Jacopini and Baeten-Boerboom’ s, but more convenient to use. Based on the new criterion, one uses an enhanced technique to show a few results which provides a deeper insight in the classification problem of λ-terms with no normal forms.
基金National Key Fundamental Research Pro-ject of China (No.2002cb312200-01-3),National Natural Science Foundation ofChina (No.60174038) and Specialized Re-search Fund for the Doctoral Program ofHigher Education (No.20030248040)
文摘The rough sets and Boolean reasoning based discretization approach (RSBRA) is no t suitable for feature selection for machine learning algorithms such as neural network or SVM because the information loss due to discretization is large. A mo dified RSBRA for feature selection was proposed and evaluated with SVM classifie rs. In the presented algorithm, the level of consistency, coined from the rough sets theory, is introduced to substitute the stop criterion of circulation of th e RSBRA, which maintains the fidelity of the training set after discretization. The experimental results show the modified algorithm has better predictive accur acy and less training time than the original RSBRA.
