Differentiable Joint Pruning and Quantization for Hardware Efficiency

By Ying Wang, Yadong Lu, Tijmen Blankevoort. 2020

94 citations recorded by the source corpus.

We present a differentiable joint pruning and quantization (DJPQ) scheme. We frame neural network compression as a joint gradient-based optimization problem, trading off between model pruning and quantization automatically for hardware efficiency. DJPQ incorporates variational information bottleneck based structured pruning and mixed-bit precision quantization into a single differentiable loss function. In contrast to previous works which consider pruning and quantization separately, our method enables users to find the optimal trade-off between both in a single training procedure. To utilize the method for more efficient hardware inference, we extend DJPQ to integrate structured pruning with power-of-two bit-restricted quantization. We show that DJPQ significantly reduces the number of Bit-Operations (BOPs) for several networks while maintaining the top-1 accuracy of original floating-point models (e.g., 53x BOPs reduction in ResNet18 on ImageNet, 43x in MobileNetV2). Compared to the conventional two-stage approach, which optimizes pruning and quantization independently, our scheme outperforms in terms of both accuracy and BOPs. Even when considering bit-restricted quantization, DJPQ achieves larger compression ratios and better accuracy than the two-stage approach.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Jianfei Cai
2. Yang He
3. Sein Park
4. Junhyuk So
5. Juncheol Shin
6. Jing Liu
7. Lingao Xiao
8. Eunhyeok Park

In Computer Vision and Pattern Recognition

1. Jianfei Cai
2. Yang He
3. Sein Park
4. Junhyuk So
5. Juncheol Shin
6. Jing Liu
7. Lingao Xiao
8. Eunhyeok Park

Citations in the corpus

• Differentiable Model Compression via Pseudo Quantization Noise (2021, neutral): “, 2021), and differentiable quantization methods (Uhlich et al., 2020; Wang et al., 2020).”
• A Survey of Quantization Methods for Efficient Neural Network Inference (2021, neutral): “It is important to consider this fact for achieving optimal benefits through hardware-aware quantization [86, 90, 238, 242, 246, 248, 257, 259].”
• Structured Pruning for Deep Convolutional Neural Networks: A Survey (2023, methodology): “Pruning and Quantization: DJPQ [162] jointly performs structured pruning and mixed-bit precision quantization in a gradient-based manner.”
• Sharpness-aware Quantization for Deep Neural Networks (2021, methodology): “Following [57, 58], we measure the computational cost of the quantized models by Bit-Operation (BOP) count for all the compared methods.”
• NIPQ: Noise Injection Pseudo Quantization for Automated DNN Optimization (2022, methodology): “This goal is achieved via various methods, e.g., RL-based [28, 29, 30], Hessian-based [31, 32, 33], and differentiable [34, 16] algorithms.”