arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.01499cs.LGmath.OCstat.ML
收藏

ReLU网络批量归一化的解密:等价凸优化模型和隐正则化

Demystifying Batch Normalization in ReLU Networks: Equivalent Convex Optimization Models and Implicit Regularization

Stanford University(斯坦福大学)

Tolga Ergen, Arda Sahiner, Batu Ozturkler, John Pauly, Morteza Mardani, Mert Pilanci

摘要:

批量归一化(BN)是一种常用的技术,用于加速和稳定深度神经网络的训练。尽管其经验的成功,一个完整的理论理解BN尚未开发。在这项工作中,我们分析BN通过凸优化的镜头。我们引入了一个基于凸对偶的分析框架,以获得具有BN的权重衰减正则化ReLU网络的精确凸表示,该网络可以在多项式时间内进行训练。我们的分析还表明,最佳层的权重可以得到简单的封闭形式的公式在高维和/或overparameterized制度。此外,我们发现,梯度下降提供了一个标准的非凸BN网络上的算法偏见的影响,我们设计了一种方法来明确编码到凸目标的隐式正则化。CIFAR图像分类的实验突出了这种显式正则化的有效性,用于模仿和大幅提高标准BN网络的性能。

英文摘要:

Batch Normalization (BN) is a commonly used technique to accelerate and stabilize training of deep neural networks. Despite its empirical success, a full theoretical understanding of BN is yet to be developed. In this work, we analyze BN through the lens of convex optimization. We introduce an analytic framework based on convex duality to obtain exact convex representations of weight-decay regularized ReLU networks with BN, which can be trained in polynomial-time. Our analyses also show that optimal layer weights can be obtained as simple closed-form formulas in the high-dimensional and/or overparameterized regimes. Furthermore, we find that Gradient Descent provides an algorithmic bias effect on the standard non-convex BN network, and we design an approach to explicitly encode this implicit regularization into the convex objective. Experiments with CIFAR image classification highlight the effectiveness of this explicit regularization for mimicking and substantially improving the performance of standard BN networks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑