用流形约束的多路连接替代残差连接，提升语音识别鲁棒性。
Beyond Residual Connections: Manifold-Constrained Hyper-Connections for Robust Speaker Representation Learning
- 将残差路径重构为多路信息混合，通过双随机矩阵实现
- 在VoxCeleb1上所有模型性能均提升，最高达0.9% EER改善
- 适合需要稳定梯度和抗信号退化的深度语音模型设计
残差连接是深度语音识别模型（如ECAPA-TDNN和ResNet）的基础，但标准恒等映射仅允许单一路径的信息流动，限制了表征能力。本文提出流形约束的超连接（mHC），将残差路径重新建模为多流演化过程，通过双随机矩阵混合信息。利用Sinkhorn-Knopp迭代，mHC保持信号强度与特征均值不变，从而稳定梯度并缓解复杂网络中的信号退化问题。我们在ECAPA-TDNN、ResNet-34、Res2Net和E-Res2Net等主干网络中替换标准残差连接，进行广泛实验。在VoxCeleb1数据集上的结果表明，mHC在所有架构中均一致提升性能，验证了其在鲁棒语音表征学习中的有效性。
原文摘要 · Abstract (English)
Residual connections are fundamental to deep speaker recogni- tion models, such as ECAPA-TDNN and ResNet. However, standard identity mapping limits information flow to a sin- gle path, constraining representation capacity. We introduce Manifold-Constrained Hyper-Connections (mHC), reformulat- ing residual paths as a multi-stream evolution where informa- tion is mixed through a doubly stochastic matrix. By employing Sinkhorn-Knopp iterations, mHC ensures energy conservation by preserving signal intensity and feature mean, which stabi- lizes gradients and mitigates signal degradation in complex net- works. We evaluate mHC by replacing standard residual con- nections in backbones including ECAPA-TDNN, ResNet-34, Res2Net, and E-Res2Net. Extensive experiments on VoxCeleb1 demonstrate that mHC connections consistently enhance per- formance across all architectures, highlighting its effectiveness for robust speaker representation learning.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。