NvidiaPatent applicationChips & hardwareInterest 3/10

Nvidia patents chip method for speeding up specific neural network math

Patent drawing
Patent drawing from the official filing.

Nvidia describes a way for its specialized processors to handle a specific type of neural network calculation called a triangular multiplicative update. The system reorganizes data in memory to match the chip's structure, allowing it to perform complex matrix math more efficiently without extra steps.

Why it mattersThis is a technical optimization for AI hardware that helps process data faster, but it is unlikely to change how everyday users interact with their devices.

In the filing’s own words

Apparatuses, systems, and techniques are to perform a triangular multiplicative update using a compiler-scheduled processor, such as a Language Processing Unit (LPU). Producer operations of a neural-network graph generate a first tensor and a second tensor from an input tensor and store the tensors in a second ordered memory layout in which a feature-channel axis precedes a first pair-index axis and a second pair-index axis. The storing performs a transpose by writing tensor elements directly to addresses corresponding to the second ordered memory layout. A triangular contraction consumes the first tensor and the second tensor in the second ordered memory layout and computes a triangular multiplicative update tensor in a batched matrix-multiplication form over an intermediate index. Downstream operations may consume the triangular multiplicative update tensor while the tensor remains stored in the second ordered memory layout.

What exactly they claim (claim 1)

1. One or more processors, comprising: circuitry to: receive an input tensor representing features and stored in memory according to a first ordered memory layout, the input tensor including a feature-channel axis, a first pair-index axis, and a second pair-index axis; execute, on a Language Processing Unit (LPU), a triangular multiplicative update graph to generate a first tensor and a second tensor and store the first tensor and the second tensor in memory according to a second ordered memory layout in which the feature-channel axis is a leading axis, wherein the storing performs a transpose that reorders the feature-channel axis relative to the first pair-index axis and the second pair-index axis; and compute, on the LPU, a triangular multiplicative update tensor by performing a tensor contraction between the first tensor and the second tensor that sums over an intermediate index.

Facts

Company
Nvidia (NVIDIA Corporation)
Official title
COMPILER-SCHEDULED TRIANGULAR UPDATE NEURAL NETWORK
Publication number
US20260359632A1
Application number
19672610
Filed
May 9, 2026
Published
Nov 12, 2026
Inventors
6
Claims
20

Classified under

G06N 3/08 Physics & computing

More from Nvidia in Chips & hardware

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.