论文标题
有效的混合网络：感应散射特征
Efficient Hybrid Network: Inducting Scattering Features
论文作者
论文摘要
最近的工作表明，将预定义和学到的过滤器结合在单个体系结构中的混合网络更适合理论分析，并且不容易在数据限制的方案中过度拟合。但是，当有足够数量的培训数据可用时，他们的表现尚未证明与传统同行的竞争。为了解决当前混合网络的核心局限性，我们引入了有效的混合网络（E-Hybridnet）。我们表明，这是第一个基于散射的方法，它在各种数据集上始终超过其常规对应物。它是通过一种新型的感应结构来实现的，该结构将散射特征嵌入了使用混合融合块中的网络流中。我们还证明了所提出的设计继承了先前的混合网络的关键特性，这是在数据限制的方案中的有效概括。我们的方法成功地结合了两个世界中最好的：学习特征的灵活性和力量，稳定性以及散射表示的可预测性。
Recent work showed that hybrid networks, which combine predefined and learnt filters within a single architecture, are more amenable to theoretical analysis and less prone to overfitting in data-limited scenarios. However, their performance has yet to prove competitive against the conventional counterparts when sufficient amounts of training data are available. In an attempt to address this core limitation of current hybrid networks, we introduce an Efficient Hybrid Network (E-HybridNet). We show that it is the first scattering based approach that consistently outperforms its conventional counterparts on a diverse range of datasets. It is achieved with a novel inductive architecture that embeds scattering features into the network flow using Hybrid Fusion Blocks. We also demonstrate that the proposed design inherits the key property of prior hybrid networks -- an effective generalisation in data-limited scenarios. Our approach successfully combines the best of the two worlds: flexibility and power of learnt features and stability and predictability of scattering representations.
