发表机构
University of Augsburg; Tensor AI Solutions GmbH; Ulm University; Centre for Advanced Analytics and Predictive Sciences, University of Augsburg(奥格斯堡大学; Tensor AI解决方案有限公司; 乌尔姆大学; 奥格斯堡大学高级分析与预测科学中心)
机构由 AI 辅助整理,请以论文原文为准。AI 中文总结
该研究针对树张量网络(TTNs)推导随机黎曼优化器,结合混合CNN-TTN架构在多数据集验证,其性能可与无约束优化媲美且支持稳定下游压缩。
AI 中文摘要
张量网络最初为量子多体物理学开发,是颇具潜力的机器学习模型。我们针对树张量网络(TTNs)的参数流形与商流形推导了随机黎曼优化器,包含适用于小批量训练的自适应及无学习率方案。采用混合CNN-TTN架构,在Fashion-MNIST、CIFAR10与Imagenette上评估这些方法,所提优化器预测性能可与无约束优化媲美,同时支持数值稳定的下游压缩。
英文摘要
Tensor networks, originally developed for quantum many-body physics, are promising models for machine learning. We derive stochastic Riemannian optimizers for tree tensor networks (TTNs) on both their parameter and quotient manifolds, including adaptive and learning-rate-free schemes suitable for minibatch training. Using a hybrid CNN-TTN architecture, we evaluate the methods on Fashion-MNIST, CIFAR10, and Imagenette. The proposed optimizers achieve predictive performance comparable to unconstrained optimization while enabling numerically stable downstream compression.
Comments26 pages, 12 figures, 5 pseudo-code algorithms; Submission to SciPost