GenQ: Quantization in Low Data Regimes with Generative Synthetic Data

arXiv:2312.05272v2 Announce Type: replace
Abstract: In the realm of deep neural network deployment, low-bit quantization presents a promising avenue for enhancing computational efficiency. However, it often hinges on the availability of training data to mitigate quantization errors, a significant challenge when data availability is scarce or restricted due to privacy or copyright concerns. Addressing this, we introduce GenQ, a novel approach employing an advanced Generative AI model to generate photorealistic, high-resolution synthetic data, overcoming the limitations of traditional methods that …

abstract arxiv availability challenge computational cs.cv data deep neural network deployment efficiency errors generative however low network neural network privacy quantization synthetic synthetic data training training data type