Deep Competitive Pathway Networks
This work addresses the challenge of improving model capability in deep learning for computer vision researchers, but it is incremental as it builds on existing grouping and residual network ideas.
The authors tackled the problem of designing deep neural architectures by proposing Competitive Pathway Networks (CoPaNet), which uses parallel residual-type subnetworks with max operations for feature competition, achieving state-of-the-art or comparable results on object recognition benchmarks like CIFAR-10, CIFAR-100, SVHN, and ImageNet.
In the design of deep neural architectures, recent studies have demonstrated the benefits of grouping subnetworks into a larger network. For examples, the Inception architecture integrates multi-scale subnetworks and the residual network can be regarded that a residual unit combines a residual subnetwork with an identity shortcut. In this work, we embrace this observation and propose the Competitive Pathway Network (CoPaNet). The CoPaNet comprises a stack of competitive pathway units and each unit contains multiple parallel residual-type subnetworks followed by a max operation for feature competition. This mechanism enhances the model capability by learning a variety of features in subnetworks. The proposed strategy explicitly shows that the features propagate through pathways in various routing patterns, which is referred to as pathway encoding of category information. Moreover, the cross-block shortcut can be added to the CoPaNet to encourage feature reuse. We evaluated the proposed CoPaNet on four object recognition benchmarks: CIFAR-10, CIFAR-100, SVHN, and ImageNet. CoPaNet obtained the state-of-the-art or comparable results using similar amounts of parameters. The code of CoPaNet is available at: https://github.com/JiaRenChang/CoPaNet.