论文标题
用多个点进行训练后量化：混合精度而没有混合精度
Post-training Quantization with Multiple Points: Mixed Precision without Mixed Precision
论文作者
论文摘要
我们考虑训练后量化问题，该问题可以离散预训练的深神经网络的权重，而不会重新训练模型。我们提出了多点量化，一种量化方法，使用低位数的多个向量的线性组合近似完整的权重矢量；这与典型的量化方法相反，该方法使用单个低精度数来近似每个权重。在计算上，我们使用有效的贪婪选择程序构建了多点量化，并根据其输出的误差自适应地决定了每个量化权重矢量的低精度点的数量。这使我们能够达到极大影响输出的重要权重的更高精度水平，从而产生“混合精度的效果”，但没有物理混合精度实现（这需要专门的硬件加速器）。从经验上讲，我们的方法可以通过通用操作数实现，几乎没有内存和计算开销。我们表明，我们的方法的表现优于Imagenet分类的一系列最新方法，并且可以将其推广到诸如Pascal VOC对象检测之类的更具挑战性的任务。
We consider the post-training quantization problem, which discretizes the weights of pre-trained deep neural networks without re-training the model. We propose multipoint quantization, a quantization method that approximates a full-precision weight vector using a linear combination of multiple vectors of low-bit numbers; this is in contrast to typical quantization methods that approximate each weight using a single low precision number. Computationally, we construct the multipoint quantization with an efficient greedy selection procedure, and adaptively decides the number of low precision points on each quantized weight vector based on the error of its output. This allows us to achieve higher precision levels for important weights that greatly influence the outputs, yielding an 'effect of mixed precision' but without physical mixed precision implementations (which requires specialized hardware accelerators). Empirically, our method can be implemented by common operands, bringing almost no memory and computation overhead. We show that our method outperforms a range of state-of-the-art methods on ImageNet classification and it can be generalized to more challenging tasks like PASCAL VOC object detection.
