Review for NeurIPS paper: Position-based Scaled Gradient for Model Quantization and Pruning

The authors confirmed that the method is a "regularization" method requiring training(and training data). In that case, I don't think the 5% degradation on W4A4 ResNet18 inference justifies the advantages claimed. The performance of PSGD is significantly worse than other quantization-aware training(QAT) work. Although the authors refer to their method as a regularization method, essentially it requires similar training data and computation as QAT. If the PTQ requires training data, then it falls back to the same level as QAT. Since PSGD requires training from scratch, not requiring an FP32 model does not seem to be an advantage to me---with training data/resources, one can always obtain an FP32 model.