Lattice Quantization

Post-training quantization of neural networks consists in quantizing a model without retraining nor hyperparameter search, while being fast and data frugal. In this paper, we propose LatticeQ, a novel post-training weight quantization method designed for deep convolutional neural networks (DCNNs). Contrary to scalar rounding widely used in state-of-the-art quantization methods, LatticeQ uses a quantizer based on lattices - discrete algebraic structures. LatticeQ exploits the inner correlations between the model parameters to the benefit of minimizing quantization error. We achieve state-of-the-art results in post-training quantization. In particular, we achieve ImageNet classification results close to full precision on Resnet-18/50, with little to no accuracy drop for 4-bit models. Our code is available here, and a more thorough version of the paper here.