2 papers
cs.LG2025
Data-Augmented Quantization-Aware Knowledge Distillation
Justin Kur, Kaiqi Zhao
Quantization-aware training (QAT) and Knowledge Distillation (KD) are combined to achieve competitive performance in creating low-bit deep learning models. Existing KD and QAT work…
cs.LG2025
Progressive Element-wise Gradient Estimation for Neural Network Quantization
Kaiqi Zhao
Neural network quantization aims to reduce the bit-widths of weights and activations, making it a critical technique for deploying deep neural networks on resource-constrained hard…