Mixed-Precision Quantization for Deep Vision Models with Integer Quadratic Programming
Quantization is a widely used technique to compress neural networks. Assigning uniform bit-widths across all layers can result in significant accuracy degradation at low precision and inefficiency at high precision. Mixed-precision quantization (MPQ) addresses this by assigning varied bit-widths to layers, optimizing the accuracy-efficiency trade-off. Existing sensitivity-based methods for MPQ assume that quantization errors across layers are independent, which leads to suboptimal choices. We introduce CLADO, a practical sensitivity-based MPQ algorithm that captures cross-layer dependency of quantization error. CLADO approximates pairwise cross-layer errors using linear equations on a small data subset. Layerwise bit-widths are assigned by optimizing a new MPQ formulation based on cross-layer quantization errors using an Integer Quadratic Program. Experiments with CNN and vision transformer models on ImageNet demonstrate that CLADO achieves state-of-the-art mixed-precision quantization performance. Code repository available here: https://github.com/JamesTuna/CLADO_MPQ
Code (1)
Tasks
QuantizationSensitivitySimilar Papers 제목 키워드 기반
HAWQV3: Dyadic Neural Network Quantization
Current low-precision quantization algorithms often have the hidden cost of conversion back and forth from floating point to quantized integer values. This hidden cost limits the latency improvement realized by quantizin…
Model CompressionQuantizationFLIQS: One-Shot Mixed-Precision Floating-Point and Integer Quantization Search
Quantization has become a mainstream compression technique for reducing model size, computational requirements, and energy consumption for modern deep neural networks (DNNs). With improved numerical support in recent har…
QuantizationMix-QSAM: Mixed-Precision Quantization of the Segment Anything Model
The Segment Anything Model (SAM) is a popular vision foundation model; however, its high computational and memory demands make deployment on resource-constrained devices challenging. While Post-Training Quantization (PTQ…
Computational EfficiencyInstance Segmentationobject-detectionObject Detection+2LampQ: Towards Accurate Layer-wise Mixed Precision Quantization for Vision Transformers
How can we accurately quantize a pre-trained Vision Transformer model? Quantization algorithms compress Vision Transformers (ViTs) into low-bit formats, reducing memory and computation demands with minimal accuracy degra…
Image ClassificationObject DetectionDQT: Dynamic Quantization Training via Dequantization-Free Nested Integer Arithmetic
The deployment of deep neural networks on resource-constrained devices relies on quantization. While static, uniform quantization applies a fixed bit-width to all inputs, it fails to adapt to their varying complexity. Dy…