Layer-wise precision optimization in analog compute-in-memory accelerators
AssigneeIntel Corporation
InventorsShamik Kundu, Arnab Raha, Richard Dorrance, Deepak Abraham Mathaikutty, Brent Carlton
Life of this patent
Estimating the term…
Loading patent…
AssigneeIntel Corporation
InventorsShamik Kundu, Arnab Raha, Richard Dorrance, Deepak Abraham Mathaikutty, Brent Carlton
Estimating the term…
It is not optimal to apply analog compute-in-memory circuitry (ACiM) for all layers of a neural network or to apply digital compute-in-memory (DCiM) circuitry for all layers of the neural network, due to the tradeoff between efficiency and precision. To address this challenge, a layer-wise offloading strategy can selectively execute neural network layers using either DCiM circuitry or ACIM circuitry based on signal and statistical sensitivity conditions. The approach leverages a combined heuristic, incorporating both the number of input channels meeting a signal sensitivity criterion and the statistical properties of the weight distribution meeting a statistical sensitivity criterion. Layers are allocated to DCiM when both conditions are satisfied, while layers are allocated to ACIM if either or both conditions are not met. The approach optimizes computational efficiency by dynamically assigning resources according to input characteristics and distributional metrics.
An apparatus for compiling a neural network model to be executed on a neural network accelerator, comprising:
One or more non-transitory computer-readable media storing instructions for compiling a neural network model to be executed on a neural network accelerator, that when executed by a processor, cause the processor to:
A method for compiling a neural network model to be executed on a neural network accelerator:
Dependencies are inferred from the claim text.