DeepAI AI Chat
Log In Sign Up

Cherry-Picking Gradients: Learning Low-Rank Embeddings of Visual Data via Differentiable Cross-Approximation

by   Mikhail Usvyatsov, et al.

We propose an end-to-end trainable framework that processes large-scale visual data tensors by looking at a fraction of their entries only. Our method combines a neural network encoder with a tensor train decomposition to learn a low-rank latent encoding, coupled with cross-approximation (CA) to learn the representation through a subset of the original samples. CA is an adaptive sampling algorithm that is native to tensor decompositions and avoids working with the full high-resolution data explicitly. Instead, it actively selects local representative samples that we fetch out-of-core and on-demand. The required number of samples grows only logarithmically with the size of the input. Our implicit representation of the tensor in the network enables processing large grids that could not be otherwise tractable in their uncompressed form. The proposed approach is particularly useful for large-scale multidimensional grid data (e.g., 3D tomography), and for tasks that require context over a large receptive field (e.g., predicting the medical condition of entire organs). The code will be available at


page 3

page 6

page 7

page 10

page 11

page 12

page 14

page 15


tntorch: Tensor Network Learning with PyTorch

We present tntorch, a tensor learning framework that supports multiple d...

Tensor Completion via a Low-Rank Approximation Pursuit

This paper considers the completion problem for a tensor (also referred ...

Fast Learnings of Coupled Nonnegative Tensor Decomposition Using Optimal Gradient and Low-rank Approximation

Nonnegative tensor decomposition has been widely applied in signal proce...

T4DT: Tensorizing Time for Learning Temporal 3D Visual Data

Unlike 2D raster images, there is no single dominant representation for ...

Tensor Robust PCA with Nonconvex and Nonlocal Regularization

Tensor robust principal component analysis (TRPCA) is a promising way fo...

Variable Bitrate Neural Fields

Neural approximations of scalar and vector fields, such as signed distan...

Compressible-composable NeRF via Rank-residual Decomposition

Neural Radiance Field (NeRF) has emerged as a compelling method to repre...