CEG4N: Counter-Example Guided Neural Network Quantization Refinement
- URL: http://arxiv.org/abs/2207.04231v1
- Date: Sat, 9 Jul 2022 09:25:45 GMT
- Title: CEG4N: Counter-Example Guided Neural Network Quantization Refinement
- Authors: Jo\~ao Batista P. Matos Jr. and Iury Bessa and Edoardo Manino and
Xidan Song and Lucas C. Cordeiro
- Abstract summary: We propose Counter-Example Guided Neural Network Quantization Refinement (CEG4N)
This technique combines search-based quantization and equivalence verification.
We produce models with up to 72% better accuracy than state-of-the-art techniques.
- Score: 2.722899166098862
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Neural networks are essential components of learning-based software systems.
However, their high compute, memory, and power requirements make using them in
low resources domains challenging. For this reason, neural networks are often
quantized before deployment. Existing quantization techniques tend to degrade
the network accuracy. We propose Counter-Example Guided Neural Network
Quantization Refinement (CEG4N). This technique combines search-based
quantization and equivalence verification: the former minimizes the
computational requirements, while the latter guarantees that the network's
output does not change after quantization. We evaluate CEG4N~on a diverse set
of benchmarks, including large and small networks. Our technique successfully
quantizes the networks in our evaluation while producing models with up to 72%
better accuracy than state-of-the-art techniques.
Related papers
Err
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.