TTML: tensor trains for general supervised machine learning
Not on the map. Why?
Abstract (arXiv)
This work proposes a novel general-purpose estimator for supervised machine learning (ML) based on tensor trains (TT). The estimator uses TTs to parametrize discretized functions, which are then optimized using Riemannian gradient descent under the form of a tensor completion problem. Since this optimization is sensitive to initialization, it turns out that the use of other ML estimators for initialization is crucial. This results in a competitive, fast ML estimator with lower memory usage than many other ML estimators, like the ones used for the initialization.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 11 of its 48 references that are in the map. First posted on arXiv 8 Mar 2022.
Builds on
11 of 48 references.
References · 11
- 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,120
- 2015Tensorizing Neural Networkscited by 246
- 2015On the Expressive Power of Deep Learning: A Tensor Analysiscited by 97
- 2017Tensor Networks for Dimensionality Reduction and Large-Scale Optimizations. Part 2 Applications and Future Perspectivescited by 53
- 2017Tensor Completion Algorithms in Big Data Analyticscited by 48
- 2021Implicit Regularization in Tensor Factorizationcited by 43
- 2018Generalized Canonical Polyadic Tensor Decompositioncited by 34
- 2017Compressing Recurrent Neural Network with Tensor Traincited by 28
- 2018Learning with tree-based tensor formatscited by 4
- 2019Nonlinear System Identification via Tensor Completioncited by 4
- 2019Tensor Completion via Gaussian Process Based Initializationcited by 3
Built on by
Citing papers per year: 2022: 1; 2023: 0; 2024: 0; 2025: 1.
2 papers.