Mixed Precision Quantization of ConvNets via Differentiable Neural Architecture Search
원문 초록
Recent work in network quantization has substantially reduced the time and space complexity of neural network inference, enabling their deployment on embedded and mobile devices with limited computational and memory resources. However, existing quantization methods often represent all weights and activations with the same precision (bit-width). In this paper, we explore a new dimension of the design space: quantizing different layers with different bit-widths. We formulate this problem as a neural architecture search problem and propose a novel differentiable neural architecture search (DNAS) framework to efficiently explore its exponential search space with gradient-based optimization. Experiments show we surpass the state-of-the-art compression of ResNet on CIFAR-10 and ImageNet. Our quantized models with 21.1x smaller model size or 103.9x lower computational cost can still outperform baseline quantized or even full precision models.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
AnalogNAS-Bench: A NAS Benchmark for Analog In-Memory Computing
From Tiny Machine Learning to Tiny Deep Learning: A Survey
Structured Pruning and Quantization for Learned Image Compression
Optimizing edge AI models on HPC systems with the edge in the loop
BASQ: Branch-wise Activation-clipping Search Quantization for Sub-4-bit Neural Networks
TAS: Ternarized Neural Architecture Search for Resource-Constrained Edge Devices
토론
이 논문과 연결된 포럼 글이 아직 없습니다.