神经元级自适应量化，边端设备高效部署。
Scale When Needed: Adaptive Neuron-level Mixed Precision Quantization Aware Training
- 每个神经元独立学习比特位数，按需动态扩展。
- 在多个模型上实现更高压缩率且精度损失小。
- 适合6G边端设备的绿色人工智能部署。
将深度神经网络部署于资源受限的6G边缘设备，需要在最小精度损失下实现激进压缩。量化感知训练（QAT）已成为主流压缩方法；然而，现有混合精度方法通常以粗粒度的层或通道为单位操作。这些方法常依赖启发式或搜索的比特分配策略，可能忽略神经元级别的细粒度差异。本文提出神经元级混合精度量化感知训练（NMP-QAT），其中每个神经元在训练过程中独立学习其离散精度。从低比特精度开始，当训练信号要求时，通过可微代理和直通估计器仅在必要时扩展比特宽度，同时保持完全离散的推理图。该适应性同时适用于权重和激活，减少内存移动。在电信与非电信数据集上的MLP和表格基础模型架构中评估，相较于混合精度QAT基线，NMP-QAT实现了更优的压缩-精度权衡，适用于网络边缘的绿色AI部署。
原文摘要 · Abstract (English)
Deploying deep neural networks on resource-constrained 6G edge devices demands aggressive compression with minimal accuracy loss. Quantization-Aware Training (QAT) has emerged as a leading compression approach; however, existing mixed-precision methods typically operate at coarse layer- or channel-level granularity. These methods often rely on heuristic or search-based bit-allocation strategies, which may overlook fine-grained variability at the neuron level. We propose Neuron-Level Mixed-Precision QAT (NMP-QAT), where each neuron independently learns its own discrete precision during training. Starting from low-bit precision, NMP-QAT expands bit-width only when training signals demand it, via differentiable surrogates and straight-through estimators, while preserving a fully discrete inference graph. This adaptability extends to both weights and activations, reducing memory movement. Evaluated on telecom and non-telecom datasets across MLP and tabular foundation model architectures, NMP-QAT achieves superior compression-accuracy trade-offs over mixed-precision QAT baselines, making it well-suited for Green AI deployments at the network edge.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。