Samsung patents method to combine neural network layers for faster chips
Samsung has filed a patent for a method that combines specific parts of a neural network into single units. The system identifies layers that can be merged to reduce the number of steps needed for the hardware to process data.
In the filing’s own words
A method of performing layer fusion on a plurality of layers, included in a neural network for execution on hardware, includes receiving an operation graph including a plurality of residual blocks and corresponding to the neural network, determining an outermost merge layer included in each of the plurality of residual blocks, determining at least one fusion receiving layer from among a plurality of outermost merge layers of the plurality of residual blocks, determining a first fused layer group including a plurality of fused layers positioned in a previous path of the at least one fusion receiving layer, and generating a fused execution graph corresponding to a first single on-chip operation unit by performing the layer fusion on the plurality of fused layers.
What exactly they claim (claim 1)
1. A method of performing layer fusion on a plurality of layers comprised in a neural network for execution on hardware, the method comprising: receiving an operation graph comprising a plurality of residual blocks and corresponding to the neural network; determining an outermost merge layer comprised in each of the plurality of residual blocks; determining at least one fusion receiving layer from among a plurality of outermost merge layers of the plurality of residual blocks; determining a first fused layer group comprising a plurality of fused layers positioned in a previous path of the at least one fusion receiving layer; and generating a fused execution graph corresponding to a first single on-chip operation unit by performing the layer fusion on the plurality of fused layers.
Facts
- Company
- Samsung (SAMSUNG ELECTRONICS CO., LTD.)
- Official title
- METHOD AND PROCESSOR FOR PERFORMING LAYER FUSION IN NEURAL NETWORK
- Publication number
- US20260268146A1
- Application number
- 19560734
- Filed
- Mar 9, 2026
- Published
- Sep 10, 2026
- Inventors
- 4
- Claims
- 20
Classified under
G06N 3/082 Physics & computingMore from Samsung in AI & software
- Samsung patents tool to erase objects from photos using AI Sep 24, 2026
- Samsung patents drawing objects into photos with AI Oct 15, 2026
- Samsung patents AI that creates personalized exercise videos Sep 10, 2026
- Samsung patents AI summaries for paused TV shows based on your viewing habits Sep 24, 2026
- Samsung patents turning chat conversations into image prompts Oct 1, 2026
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.