Method for determining quantization parameters in neural networks and related products
Patent Information
- Application Number
- JP2020573207
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2019-06-27
- Filing Date
- 2019-09-19
- Publication Date
- 2024-11-06
- Estimated Expiration
- 2039-09-19
AI Technical Summary
Neural networks require high-precision data formats like floating-point numbers, which occupy significant storage space and increase processing speed and bandwidth requirements, leading to higher costs and resource consumption.
A method and device for adjusting data bit width through quantization, converting high-precision data to low-precision fixed-point numbers, allowing neural network operations to reduce storage space and improve computing performance by determining optimal bit widths based on quantization error and statistical data analysis.
This approach reduces the space required for neural network storage and improves computing efficiency by converting data to fixed-point numbers, maintaining accuracy while minimizing resource consumption.
Smart Images

Figure 0007578493000043 
Figure 0007578493000044 
Figure 0007578493000045
Abstract
Citation Information
Patent Citations
Encoding system and decoding system
JP1998233691A
Adaptive bit-width reduction for neural networks
US20190050710A1
