Neural Network Weight Accumulation for Low-Precision Training

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing neural networks face challenges in reducing computational amount while minimizing precision loss, particularly when using low-precision number systems for training and updating weights.

Innovation Solution

A method and apparatus for neural network training that involves calculating individual update values, accumulating them in a buffer, and updating weights only when the accumulated update value exceeds a threshold, which corresponds to the least significant effective bit of the weight, while adjusting precision through decimal point positioning and bias adjustments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If low-precision number systems are used for training neural networks, then computational amount is reduced, but precision loss occurs

Engineering Contradiction:
Improvecomputational amountVSAvoidprecision loss
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent merges multiple small update values into an accumulated update value by summing them together before applying the weight update. This accumulation buffer stores the sum of gradients from multiple training iterations, allowing low-precision individual updates to be combined into a more precise cumulative update that reduces overall precision loss while maintaining computational efficiency.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent performs preliminary accumulation of update values in an accumulation buffer before actually updating the weight. By pre-summing multiple small updates and only applying the update when the accumulated value exceeds a threshold, the system prepares precise update values in advance, minimizing precision loss during the actual weight update operation.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If individual weight updates are performed frequently, then training accuracy is improved, but computational overhead increases

Engineering Contradiction:
Improvetraining accuracyVSAvoidcomputational overhead
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements periodic weight updates by accumulating update values over multiple training iterations and only applying the update when the accumulated value exceeds a threshold. This periodic update mechanism reduces computational overhead by avoiding frequent weight updates while maintaining training accuracy through the accumulation of meaningful gradient information.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The patent uses a threshold-based feedback mechanism that monitors the accumulated update value and only triggers a weight update when the accumulation exceeds the threshold. This feedback control ensures that updates are performed only when sufficiently meaningful, balancing training accuracy with computational efficiency by preventing unnecessary update operations.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3474195B1Method and apparatus with neural network
Publication Date: 2025.07.16 SAMSUNG ELECTRONICS CO LTD
  • EP3474195B1 patent drawingFigure 1
  • EP3474195B1 patent drawingFigure 2
  • EP3474195B1 patent drawingFigure 3A~3B

AI summary

A processor-implemented neural network method includes calculating individual update values for a weight assigned to a connection relationship between nodes included in a neural network; generating an accumulated update value by accumulating the individual update values in an accumulation buffer; and training the neural network by updating the weight using the accumulated update value in response to the accumulated update value being equal to or greater than a threshold value.