BalancedTNN

Tensor Completion using OTNN and LTNN

This repository implements tensor completion methods using Overlapped Tensor Nuclear Norm (OTNN) and Latent Tensor Nuclear Norm (LTNN) algorithms [1]. These methods are designed to recover missing entries in multi-dimensional data (tensors).

[1] Qiu, Y., Zhou, G., Wang, A., Zhao, Q., & Xie, S. (2024). Balanced Unfolding Induced Tensor Nuclear Norms for High-Order Tensor Completion. IEEE Transactions on Neural Networks and Learning Systems, early access.

Requirements

• Tensor Toolbox in main dir (for some auxiliary functions),

Citation

If you find this repository useful for your research, please cite the following paper:

@article{qiu2024balanced,
  title={Balanced Unfolding Induced Tensor Nuclear Norms for High-Order Tensor Completion},
  author={Qiu, Yuning and Zhou, Guoxu and Wang, Andong and Zhao, Qibin and Xie, Shengli},
  journal={IEEE Transactions on Neural Networks and Learning Systems},
  year={2024},
  publisher={IEEE}
}