Feature selection for classification applications in neural network pruning by Gong, Rui

“…Then the neural network pruning problem is formulated as a feature selection problems, and feature selection methods are introduced to prune neural networks. …”
Get full text

Unsupervised generative variational continual learning by Liu, Guimeng

“…Recent research in continual learning generally incorporates two of these methods to obtain better performance. …”
Get full text

Refining learning models in grammatical inference by Wang, Xiangrui

“…We introduce the use of recurrent neural networks (RNNs) and present a pruning learning method to avoid the exponential space costs. …”
Get full text

Towards ILP-based LTLf passive learning by Ielo, A, Law, M, Fionda, V, Ricca, F, De Giacomo, G, De Giacomo, G, Russo, A

“…Inferring a LTL_f formula from a set of example traces, also known as passive learning, is a challenging task for model-based techniques. …”

Learning class-specific edges for object detection and segmentation by Prasad, M, Zisserman, A, Fitzgibbon, A, Kumar, MP, Torr, PHS

“…However, for the most part, these edge-based methods operate solely on the geometric shape of edges, treating them equally and ignoring the fact that for certain object classes, the appearance of the object on the “inside” of the edge may provide valuable recognition cues.…”

Red Alarm for Pre-trained Models: Universal Vulnerability to Neuron-level Backdoor Attacks by Zhang, Zhengyan, Xiao, Guangxuan, Li, Yongwei, Lv, Tian, Qi, Fanchao, Liu, Zhiyuan, Wang, Yasheng, Jiang, Xin, Sun, Maosong

“…Finally, we apply several defense methods to NeuBA and find that model pruning is a promising technique to resist NeuBA by omitting backdoored neurons.…”
Get full text

Efficient algorithms for subgraph counting and enumeration on large graphs by Wang, Kaixin

“…It determines the weights of edges in a data-driven fashion, using a novel method based on reinforcement learning. We conduct extensive experi- ments to verify that our technique can produce estimates with smaller errors while often running faster compared with existing algorithms. …”
Get full text