Research map / Learning theory and algorithms / Matrix and tensor methods
Legendre Decomposition for Tensors
Mahito Sugiyama, Hiroyuki Nakahara, Koji Tsuda
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use natural gradient instead of gradient descent | Legendre decomposition · Synthetic 20 × 20 × 20 tensor | running time: more than six orders of magnitude faster | Improved |
| Use natural gradient instead of gradient descent | Legendre decomposition · Synthetic 20 × 20 × 20 tensor | iterations to convergence: two or three iterations; more than 10^5 iterations | Improved |
| Use Legendre decomposition instead of nonnegative Tucker, nonnegative CP, or CP-APR | Tensor decomposition · Face image dataset | RMSE | Mixed |
| Use Legendre decomposition instead of nonnegative Tucker, nonnegative CP, or CP-APR | Tensor decomposition · MNIST | RMSE | Improved |
| Remove zero entries from the sample space before decomposition | Legendre decomposition · MNIST | RMSE | Improved |
| Use Legendre decomposition instead of Tucker and CP decompositions | Tensor decomposition · Face image dataset | running time | Hurt |
| Use Legendre decomposition instead of Tucker and CP decompositions | Tensor decomposition · MNIST | running time | Hurt |
Closest papers by meaning
- Expected Tensor Decomposition with Stochastic Gradient Descent (AAAI 2016 · similarity 0.85)
- Provable Online CP/PARAFAC Decomposition of a Structured Tensor via Dictionary Learning (NeurIPS 2020 · similarity 0.85)
- Nonparametric Decomposition of Sparse Tensors (ICML 2021 · similarity 0.85)
- Doubly Decomposing Nonparametric Tensor Regression (ICML 2016 · similarity 0.84)
- On Tensor Train Rank Minimization : Statistical Efficiency and Scalable Algorithm (NeurIPS 2017 · similarity 0.84)
- Mode-wise Tensor Decompositions: Multi-dimensional Generalizations of CUR Decompositions (JMLR 2021 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
