Composite nonlinear feedback （CNF） control techniquefor tracking control problems is extended to the output regulationproblem of singular linear systems with input saturation. A statefeedback CNF control law and an ...Composite nonlinear feedback （CNF） control techniquefor tracking control problems is extended to the output regulationproblem of singular linear systems with input saturation. A statefeedback CNF control law and an output feedback CNF controllaw are constructed respectively for the output regulation problemof singular linear systems with input saturation. It is shown thatthe output regulation problem by CNF control is solvable underthe same solvability conditions of the output regulation problemby linear control. However, with the virtue of the CNF control, thetransient performance of the closed-loop system can be improvedby carefully designing the linear part and the nonlinear part of theCNF control law. The design procedure and the improvement ofthe transient performance of the closed-loop system are illustratedwith a numerical simulation.展开更多
Structural neural network pruning aims to remove the redundant channels in the deep convolutional neural networks(CNNs)by pruning the filters of less importance to the final output accuracy.To reduce the degradation o...Structural neural network pruning aims to remove the redundant channels in the deep convolutional neural networks(CNNs)by pruning the filters of less importance to the final output accuracy.To reduce the degradation of performance after pruning,many methods utilize the loss with sparse regularization to produce structured sparsity.In this paper,we analyze these sparsity-training-based methods and find that the regularization of unpruned channels is unnecessary.Moreover,it restricts the network′s capacity,which leads to under-fitting.To solve this problem,we propose a novel pruning method,named Mask Sparsity,with pruning-aware sparse regularization.Mask Sparsity imposes the fine-grained sparse regularization on the specific filters selected by a pruning mask,rather than all the filters of the model.Before the fine-grained sparse regularization of Mask Sparity,we can use many methods to get the pruning mask,such as running the global sparse regularization.Mask Sparsity achieves a 63.03%float point operations(FLOPs)reduction on Res Net-110 by removing 60.34%of the parameters,with no top-1 accuracy loss on CIFAR-10.On ILSVRC-2012,Mask Sparsity reduces more than 51.07%FLOPs on Res Net-50,with only a loss of 0.76%in the top-1 accuracy.The code of this paper is released at https://github.com/CASIA-IVA-Lab/Mask Sparsity.We have also integrated the code into a self-developed Py Torch pruning toolkit,named Easy Pruner,at https://gitee.com/casia_iva_engineer/easypruner.展开更多
Objective To study the influence of omentin on the expression of collagenⅠ/Ⅳin the human aortic smooth muscle cells subjected to oxidative stress.Methods Human aortic smooth muscle cells were cultured in vitro.When ...Objective To study the influence of omentin on the expression of collagenⅠ/Ⅳin the human aortic smooth muscle cells subjected to oxidative stress.Methods Human aortic smooth muscle cells were cultured in vitro.When cell density reached above 90%,cells were randomly divided into 5 groups:control group。展开更多
Objective To explore the prognostic value of regulatory T cells (Tregs) and M2 macrophages in diffuse large B-cell lymphoma (DLBCL) tissues.Methods The expression of CD163 and Foxp3 was detected by immunohistochemistr...Objective To explore the prognostic value of regulatory T cells (Tregs) and M2 macrophages in diffuse large B-cell lymphoma (DLBCL) tissues.Methods The expression of CD163 and Foxp3 was detected by immunohistochemistry in 92 cases of DLBCL,and it展开更多
Myxococcus xanthus provides an excellent model organism for studying the mechanism of multicellular morphogenesis. The mRNA for FruA, a transcription factor essential for the development of M. xanthus, contains a very...Myxococcus xanthus provides an excellent model organism for studying the mechanism of multicellular morphogenesis. The mRNA for FruA, a transcription factor essential for the development of M. xanthus, contains a very long 5′-UTR consisting of 235 nucleotides. Using lacZ as a reporter gene, two fruA-lacZ translational fusions retaining or lacking the fruA 5′-UTR were constructed and separately integrated at phage Mx8 attachment site (attB) in M. xanthus chromosome. Deletion in 5′-UTR between nucleotides from +4 to +220 abolished fruA-lacZ expression during develop- ment, indicating that the 5′-UTR is essential for the induction of fruA. Prediction of the RNA secondary structure of 5′-UTR shows that this region could form an extremely sta- ble three-helix junction structure, which might be a binding site for a regulatory protein or contain a cis-acting element(s) to control fruA expression. Thus, the 5′-UTR of fruA mRNA positively regulates the expression of its own gene.展开更多
基金supported by the National Natural Science Foundation of China(61374035)
文摘Composite nonlinear feedback （CNF） control techniquefor tracking control problems is extended to the output regulationproblem of singular linear systems with input saturation. A statefeedback CNF control law and an output feedback CNF controllaw are constructed respectively for the output regulation problemof singular linear systems with input saturation. It is shown thatthe output regulation problem by CNF control is solvable underthe same solvability conditions of the output regulation problemby linear control. However, with the virtue of the CNF control, thetransient performance of the closed-loop system can be improvedby carefully designing the linear part and the nonlinear part of theCNF control law. The design procedure and the improvement ofthe transient performance of the closed-loop system are illustratedwith a numerical simulation.
基金supported by National Natural Science Foundation of China(Nos.62176254,61976210,61876086,62076235,62002356,62006230 and 62002357)National Key R&D Program of China(No.2021ZD0110403).
文摘Structural neural network pruning aims to remove the redundant channels in the deep convolutional neural networks(CNNs)by pruning the filters of less importance to the final output accuracy.To reduce the degradation of performance after pruning,many methods utilize the loss with sparse regularization to produce structured sparsity.In this paper,we analyze these sparsity-training-based methods and find that the regularization of unpruned channels is unnecessary.Moreover,it restricts the network′s capacity,which leads to under-fitting.To solve this problem,we propose a novel pruning method,named Mask Sparsity,with pruning-aware sparse regularization.Mask Sparsity imposes the fine-grained sparse regularization on the specific filters selected by a pruning mask,rather than all the filters of the model.Before the fine-grained sparse regularization of Mask Sparity,we can use many methods to get the pruning mask,such as running the global sparse regularization.Mask Sparsity achieves a 63.03%float point operations(FLOPs)reduction on Res Net-110 by removing 60.34%of the parameters,with no top-1 accuracy loss on CIFAR-10.On ILSVRC-2012,Mask Sparsity reduces more than 51.07%FLOPs on Res Net-50,with only a loss of 0.76%in the top-1 accuracy.The code of this paper is released at https://github.com/CASIA-IVA-Lab/Mask Sparsity.We have also integrated the code into a self-developed Py Torch pruning toolkit,named Easy Pruner,at https://gitee.com/casia_iva_engineer/easypruner.
文摘Objective To study the influence of omentin on the expression of collagenⅠ/Ⅳin the human aortic smooth muscle cells subjected to oxidative stress.Methods Human aortic smooth muscle cells were cultured in vitro.When cell density reached above 90%,cells were randomly divided into 5 groups:control group。
文摘Objective To explore the prognostic value of regulatory T cells (Tregs) and M2 macrophages in diffuse large B-cell lymphoma (DLBCL) tissues.Methods The expression of CD163 and Foxp3 was detected by immunohistochemistry in 92 cases of DLBCL,and it
文摘Myxococcus xanthus provides an excellent model organism for studying the mechanism of multicellular morphogenesis. The mRNA for FruA, a transcription factor essential for the development of M. xanthus, contains a very long 5′-UTR consisting of 235 nucleotides. Using lacZ as a reporter gene, two fruA-lacZ translational fusions retaining or lacking the fruA 5′-UTR were constructed and separately integrated at phage Mx8 attachment site (attB) in M. xanthus chromosome. Deletion in 5′-UTR between nucleotides from +4 to +220 abolished fruA-lacZ expression during develop- ment, indicating that the 5′-UTR is essential for the induction of fruA. Prediction of the RNA secondary structure of 5′-UTR shows that this region could form an extremely sta- ble three-helix junction structure, which might be a binding site for a regulatory protein or contain a cis-acting element(s) to control fruA expression. Thus, the 5′-UTR of fruA mRNA positively regulates the expression of its own gene.
