The filled function algorithm is an important method to solve global optimization problems.In this paper,a parameter-free filled function is proposed for solving general global optimization problem,discuss the theoret...The filled function algorithm is an important method to solve global optimization problems.In this paper,a parameter-free filled function is proposed for solving general global optimization problem,discuss the theoretical properties of this function and give the corresponding algorithm.The numerical experiments on some typical test problems using the algorithm and the numerical results show that the algorithm is effective.Applying the filled function method to the parameter solving problem in the logical population growth model,and then can be effectively applied to Chinese population prediction.The experimental results show that the algorithm has good practicability in practical application.展开更多
Nowadays,the upwind schemes are in a rapid development to capture shock accurately.However,these upwind schemes’properties at low speeds,such as their reconstruction scheme dependencies,grid dependencies,and Mach num...Nowadays,the upwind schemes are in a rapid development to capture shock accurately.However,these upwind schemes’properties at low speeds,such as their reconstruction scheme dependencies,grid dependencies,and Mach number dependencies,are concerned by few people.In this paper,a systematic study on their low speeds’issues is conducted.Through a series of tests,we can find that most parameter-free upwind schemes,widely used in practice today,are not applicable to low speeds’simulations.In contrast,SLAU and SLAU2 can give reliable results.Also,the upwind scheme’s influence on the accuracy is stronger than the reconstruction scheme’s influence at low speeds.展开更多
Time-series discord is widely used in data mining applications to characterize anomalous subsequences in time series. Compared to some other discord search algorithms, the direct search algorithm based on the recurren...Time-series discord is widely used in data mining applications to characterize anomalous subsequences in time series. Compared to some other discord search algorithms, the direct search algorithm based on the recurrence plot shows the advantage of being fast and parameter free. The direct search algorithm, however, relies on quasi-periodicity in input time series, an assumption that limits the algorithm＇s applicability. In this paper, we eliminate the periodicity assumption from the direct search algorithm by proposing a reference function for subsequences and a new sampling strategy based on the reference function. These measures result in a new algorithm with improved efficiency and robustness, as evidenced by our empirical evaluation.展开更多
Knowledge distillation aims to distill knowledge from teacher networks to train student networks.Distilling intermediate features has attracted much attention in recent years as it can beflexibly applied in variousfiel...Knowledge distillation aims to distill knowledge from teacher networks to train student networks.Distilling intermediate features has attracted much attention in recent years as it can beflexibly applied in variousfields such as image classification,object detection and semantic segmentation.A critical obstacle of feature-based knowledge distillation is the dimension gap between the intermediate features of teacher and student,and plenty of methods have been proposed to resolve this problem.However,these works usually implement feature uniformization in an unsupervised way,lacking guidance to help the student network learn meaningful mapping functions in the uniformization process.Moreover,the dimension uniformization process of the student and teacher network is usually not equivalent as the mapping functions are different.To this end,some factors of the feature are discarded during parametric feature alignment,or some factors are blended in some non-parametric operations.In this paper,we propose a novel semantic-aware knowledge distillation scheme to solve these problems.We build a standalone feature-based classification branch to extract semantic-aware knowledge for better guiding the learning process of the student network.In addition,to avoid the inequivalence of feature uniformization between teacher and student,we design a novel parameter-free self-attention operation that can convert features of different dimensions into vectors of the same length.Experimental results show that the proposed knowledge distillation scheme outperforms existing feature-based distillation methods on the widely used CIFAR-100 and CINIC-10 datasets.展开更多
In response to the scarcity of infrared aircraft samples and the tendency of traditional deep learning to overfit,a few-shot infrared aircraft classification method based on cross-correlation networks is proposed.This...In response to the scarcity of infrared aircraft samples and the tendency of traditional deep learning to overfit,a few-shot infrared aircraft classification method based on cross-correlation networks is proposed.This method combines two core modules:a simple parameter-free self-attention and cross-attention.By analyzing the self-correlation and cross-correlation between support images and query images,it achieves effective classification of infrared aircraft under few-shot conditions.The proposed cross-correlation network integrates these two modules and is trained in an end-to-end manner.The simple parameter-free self-attention is responsible for extracting the internal structure of the image while the cross-attention can calculate the cross-correlation between images further extracting and fusing the features between images.Compared with existing few-shot infrared target classification models,this model focuses on the geometric structure and thermal texture information of infrared images by modeling the semantic relevance between the features of the support set and query set,thus better attending to the target objects.Experimental results show that this method outperforms existing infrared aircraft classification methods in various classification tasks,with the highest classification accuracy improvement exceeding 3%.In addition,ablation experiments and comparative experiments also prove the effectiveness of the method.展开更多
基金Supported by National Natural Science Foundation of China(Grant No.12071112,11471102)Basic Research Projects for Key Scientic Research Projects in Henan Province(Grant No.20ZX001).
文摘The filled function algorithm is an important method to solve global optimization problems.In this paper,a parameter-free filled function is proposed for solving general global optimization problem,discuss the theoretical properties of this function and give the corresponding algorithm.The numerical experiments on some typical test problems using the algorithm and the numerical results show that the algorithm is effective.Applying the filled function method to the parameter solving problem in the logical population growth model,and then can be effectively applied to Chinese population prediction.The experimental results show that the algorithm has good practicability in practical application.
基金supported by the National Basic Research Program of China("973" Project)(Grant No.2009CB724104)
文摘Nowadays,the upwind schemes are in a rapid development to capture shock accurately.However,these upwind schemes’properties at low speeds,such as their reconstruction scheme dependencies,grid dependencies,and Mach number dependencies,are concerned by few people.In this paper,a systematic study on their low speeds’issues is conducted.Through a series of tests,we can find that most parameter-free upwind schemes,widely used in practice today,are not applicable to low speeds’simulations.In contrast,SLAU and SLAU2 can give reliable results.Also,the upwind scheme’s influence on the accuracy is stronger than the reconstruction scheme’s influence at low speeds.
基金Support by Australian Research Council Linkage Grant No. LP 0776417
文摘Time-series discord is widely used in data mining applications to characterize anomalous subsequences in time series. Compared to some other discord search algorithms, the direct search algorithm based on the recurrence plot shows the advantage of being fast and parameter free. The direct search algorithm, however, relies on quasi-periodicity in input time series, an assumption that limits the algorithm＇s applicability. In this paper, we eliminate the periodicity assumption from the direct search algorithm by proposing a reference function for subsequences and a new sampling strategy based on the reference function. These measures result in a new algorithm with improved efficiency and robustness, as evidenced by our empirical evaluation.
基金supported in part by Key-Area Research and Development Program of Guangdong Province(Grant No.2021B0101200001)the National Natural Science Foundation of China(Grant No.62293543)+1 种基金the National Natural Science Foundation of China(Grant No.U21B2048)the Open Research Projects of Zhejiang Lab(Grant No.2019KD0AD01/010).
文摘Knowledge distillation aims to distill knowledge from teacher networks to train student networks.Distilling intermediate features has attracted much attention in recent years as it can beflexibly applied in variousfields such as image classification,object detection and semantic segmentation.A critical obstacle of feature-based knowledge distillation is the dimension gap between the intermediate features of teacher and student,and plenty of methods have been proposed to resolve this problem.However,these works usually implement feature uniformization in an unsupervised way,lacking guidance to help the student network learn meaningful mapping functions in the uniformization process.Moreover,the dimension uniformization process of the student and teacher network is usually not equivalent as the mapping functions are different.To this end,some factors of the feature are discarded during parametric feature alignment,or some factors are blended in some non-parametric operations.In this paper,we propose a novel semantic-aware knowledge distillation scheme to solve these problems.We build a standalone feature-based classification branch to extract semantic-aware knowledge for better guiding the learning process of the student network.In addition,to avoid the inequivalence of feature uniformization between teacher and student,we design a novel parameter-free self-attention operation that can convert features of different dimensions into vectors of the same length.Experimental results show that the proposed knowledge distillation scheme outperforms existing feature-based distillation methods on the widely used CIFAR-100 and CINIC-10 datasets.
基金Supported by the National Pre-research Program during the 14th Five-Year Plan(514010405)。
文摘In response to the scarcity of infrared aircraft samples and the tendency of traditional deep learning to overfit,a few-shot infrared aircraft classification method based on cross-correlation networks is proposed.This method combines two core modules:a simple parameter-free self-attention and cross-attention.By analyzing the self-correlation and cross-correlation between support images and query images,it achieves effective classification of infrared aircraft under few-shot conditions.The proposed cross-correlation network integrates these two modules and is trained in an end-to-end manner.The simple parameter-free self-attention is responsible for extracting the internal structure of the image while the cross-attention can calculate the cross-correlation between images further extracting and fusing the features between images.Compared with existing few-shot infrared target classification models,this model focuses on the geometric structure and thermal texture information of infrared images by modeling the semantic relevance between the features of the support set and query set,thus better attending to the target objects.Experimental results show that this method outperforms existing infrared aircraft classification methods in various classification tasks,with the highest classification accuracy improvement exceeding 3%.In addition,ablation experiments and comparative experiments also prove the effectiveness of the method.
基金Supported by the National Basic Research Program of China under Grant No.2002CB312101(国家重点基础研究发展计划(973))the National Natural Science Foundation of China under Grant Nos.60373036,60333010(国家自然科学基金)+1 种基金the Doctoral Program of Higher Education(Specialized Research Fund)of China under Grant No.20050335069(国家教育部高等学校博士学科点专项科研基金)the Natural Science Foundation of Zhejiang Province of China under Grant No.R106449(浙江省自然科学基金)
