Dynamic DropConnect: Enhancing Neural Network Robustness through Adaptive Edge Dropping Strategies

TL;DR

This paper proposes a novel dynamic DropConnect method that assigns adaptive drop rates to edges in neural networks, improving robustness and generalization without added complexity, validated through experiments on various datasets.

Contribution

Introduces a dynamic, edge-specific DropConnect technique that adapts drop rates based on gradient magnitudes, enhancing neural network robustness and performance.

Findings

Outperforms traditional Dropout, DropConnect, and Standout methods.

Improves robustness and generalization without increasing computational cost.

Validated on synthetic and real datasets.

Abstract

Dropout and DropConnect are well-known techniques that apply a consistent drop rate to randomly deactivate neurons or edges in a neural network layer during training. This paper introduces a novel methodology that assigns dynamic drop rates to each edge within a layer, uniquely tailoring the dropping process without incorporating additional learning parameters. We perform experiments on synthetic and openly available datasets to validate the effectiveness of our approach. The results demonstrate that our method outperforms Dropout, DropConnect, and Standout, a classic mechanism known for its adaptive dropout capabilities. Furthermore, our approach improves the robustness and generalization of neural network training without increasing computational complexity. The complete implementation of our methodology is publicly accessible for research and replication purposes at https://github.com/ericabd888/Adjusting-the-drop-probability-in-DropConnect-based-on-the-magnitude-of-the-gradient/.