Feature-Dependent Cross-Connections in Multi-Path Neural Networks

TL;DR

This paper introduces feature-dependent cross-connections in multi-path neural networks, enabling context-specific feature learning and reducing redundancy, which improves image recognition accuracy without increasing complexity.

Contribution

It proposes a novel mechanism for feature-dependent cross-connections that adaptively allocate resources in multi-path networks, enhancing specialization and efficiency.

Findings

Improved image recognition accuracy on multiple datasets.

Achieves better performance at similar complexity levels.

Reduces feature redundancy through adaptive resource allocation.

Abstract

Learning a particular task from a dataset, samples in which originate from diverse contexts, is challenging, and usually addressed by deepening or widening standard neural networks. As opposed to conventional network widening, multi-path architectures restrict the quadratic increment of complexity to a linear scale. However, existing multi-column/path networks or model ensembling methods do not consider any feature-dependent allocation of parallel resources, and therefore, tend to learn redundant features. Given a layer in a multi-path network, if we restrict each path to learn a context-specific set of features and introduce a mechanism to intelligently allocate incoming feature maps to such paths, each path can specialize in a certain context, reducing the redundancy and improving the quality of extracted features. This eventually leads to better-optimized usage of parallel resources. To do this, we propose inserting feature-dependent cross-connections between parallel sets of feature maps in successive layers. The weighting coefficients of these cross-connections are computed from the input features of the particular layer. Our multi-path networks show improved image recognition accuracy at a similar complexity compared to conventional and state-of-the-art methods for deepening, widening and adaptive feature extracting, in both small and large scale datasets.