TensorKrowch: Smooth integration of tensor networks in machine learning
This work provides a tool for researchers and practitioners in machine learning to more easily adopt tensor network architectures, though it is incremental as it focuses on software development rather than novel algorithmic advances.
The authors tackled the challenge of integrating tensor networks into machine learning pipelines by introducing TensorKrowch, an open-source Python library built on PyTorch, which enables users to construct, train, and incorporate tensor networks as layers in complex deep learning models.
Tensor networks are factorizations of high-dimensional tensors into networks of smaller tensors. They have applications in physics and mathematics, and recently have been proposed as promising machine learning architectures. To ease the integration of tensor networks in machine learning pipelines, we introduce TensorKrowch, an open source Python library built on top of PyTorch. Providing a user-friendly interface, TensorKrowch allows users to construct any tensor network, train it, and integrate it as a layer in more intricate deep learning models. In this paper, we describe the main functionality and basic usage of TensorKrowch, and provide technical details on its building blocks and the optimizations performed to achieve efficient operation.