arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2009.02191cs.LGcs.CV
收藏

双精度深度神经网络

Dual Precision Deep Neural Network

Sungkyunkwan University(成均馆大学)

Jae Hyun Park, Ji Sub Choi, Jong Hwan Ko

AI总结:

针对深度神经网络推理时精度与复杂度需在线权衡的问题,提出单模型含双精度模式、支持免重训练在线切换的双精度DNN,通过两阶段训练同时优化高低精度模式。

摘要:

深度神经网络(DNNs)的在线精度可扩展性是在DNN推理期间支持精度与复杂度权衡的关键特性。本文提出了双精度DNN,在单个模型中包含两种不同的精度模式,从而无需重新训练即可支持在线精度切换。所提出的两阶段训练过程同时优化低精度和高精度模式。

英文摘要:

On-line Precision scalability of the deep neural networks(DNNs) is a critical feature to support accuracy and complexity trade-off during the DNN inference. In this paper, we propose dual-precision DNN that includes two different precision modes in a single model, thereby supporting an on-line precision switch without re-training. The proposed two-phase training process optimizes both low- and high-precision modes.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑