Lossy tensor compression method using neural network-based tensor-train decomposition
AssigneesKorean Advanced Institute of Science and Technology, Foundation of Soongsil University-Industry Cooperation
InventorsKijung Shin, Taehyung Kwon, Jihoon Ko, Jinhong Jung
Life of this patent
Estimating the term…
Abstract
Disclosed is a lossy tensor compression method using neural tensor-train decomposition (NTTD). A lossy tensor compression method performed by a computer system may include inputting, to a neural tensor-train decomposition (NTTD) model, mode indices of a target entry to be reconstructed, and obtaining tensor-train (TT) cores from the mode indices of the target entry to be reconstructed through the NTTD model.
Claims10
A lossy tensor compression method performed by a computer system, comprising:
- inputting, to a neural tensor-train decomposition (NTTD) model, mode indices of a target entry to be reconstructed;
- and
- obtaining tensor-train (TT) cores from the mode indices of the target entry to be reconstructed through the NTTD model.
A computer-readable recording medium in which a computer program for executing a lossy tensor compression method performed by a computer system has been stored,
- wherein the lossy tensor compression method comprises:
- inputting, to a neural tensor-train decomposition (NTTD) model, mode indices of a target entry to be reconstructed;
- and
- obtaining tensor-train (TT) cores from the mode indices of the target entry to be reconstructed through the NTTD model.
A computer system comprising:
- memory;
- and
- a processor connected to the memory and configured to execute at least one instruction stored in the memory,
- wherein the processor inputs, to a neural tensor-train decomposition (NTTD) model, mode indices of a target entry to be reconstructed and obtains tensor-train (TT) cores from the mode indices of the target entry to be reconstructed through the NTTD model.
Dependencies are inferred from the claim text.