Batch Normalization-Free Fully Integer Quantized Neural Networks via Progressive Tandem Learning
By: Pengfei Sun , Wenyu Jiang , Piew Yoong Chee and more
Quantised neural networks (QNNs) shrink models and reduce inference energy through low-bit arithmetic, yet most still depend on a running statistics batch normalisation (BN) layer, preventing true integer-only deployment. Prior attempts remove BN by parameter folding or tailored initialisation; while helpful, they rarely recover BN's stability and accuracy and often impose bespoke constraints. We present a BN-free, fully integer QNN trained via a progressive, layer-wise distillation scheme that slots into existing low-bit pipelines. Starting from a pretrained BN-enabled teacher, we use layer-wise targets and progressive compensation to train a student that performs inference exclusively with integer arithmetic and contains no BN operations. On ImageNet with AlexNet, the BN-free model attains competitive Top-1 accuracy under aggressive quantisation. The procedure integrates directly with standard quantisation workflows, enabling end-to-end integer-only inference for resource-constrained settings such as edge and embedded devices.
Similar Papers
DQT: Dynamic Quantization Training via Dequantization-Free Nested Integer Arithmetic
Machine Learning (CS)
Makes AI smarter using less computer power.
Uncertainty-Preserving QBNNs: Multi-Level Quantization of SVI-Based Bayesian Neural Networks for Image Classification
Machine Learning (CS)
Makes smart computers work with less power.
Rescaling-Aware Training for Efficient Deployment of Deep Learning Models on Full-Integer Hardware
Machine Learning (CS)
Makes AI on small devices run faster, cheaper.