US 12,732,207 B2
Lossy tensor compression method using neural network-based tensor-train decomposition
Kijung Shin, Daejeon (KR); Taehyung Kwon, Daejeon (KR); Jihoon Ko, Daejeon (KR); and Jinhong Jung, Seoul (KR)
Assigned to Korean Advanced Institute of Science and Technology, Daejeon (KR); and Foundation of Soongsil University-Industry Cooperation, Seoul (KR)
Filed by Korea Advanced Institute of Science and Technology, Daejeon (KR); and Foundation of Soongsil University-Industry Cooperation, Seoul (KR)
Filed on Oct. 2, 2024, as Appl. No. 18/904,499.
Claims priority of application No. 10-2024-0010698 (KR), filed on Jan. 24, 2024.
Prior Publication US 2025/0240028 A1, Jul. 24, 2025
Int. Cl. H03M 7/30 (2006.01); G06N 3/0442 (2023.01); G06N 3/0495 (2023.01)
CPC H03M 7/3059 (2013.01) [H03M 7/3077 (2013.01); H03M 7/6011 (2013.01); G06N 3/0442 (2023.01); G06N 3/0495 (2023.01)] 10 Claims
OG exemplary drawing
 
9. 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.