Auto Deep Compression by Reinforcement Learning Based Actor-Critic Structure
Hamed Hakkak
文章地址
摘要
基于模型的壓縮是一種有效的、方便的、擴(kuò)展的神經(jīng)網(wǎng)絡(luò)模型模型，它需要的計(jì)算能力和功耗都是有限的。然而，傳統(tǒng)的壓縮技術(shù)模型使用了精心設(shè)計(jì)的特性，并探索了在大小、速度和精度方面的大型空間的探索和設(shè)計(jì)的專門領(lǐng)域，這些領(lǐng)域通常的回報(bào)更少，時(shí)間也在增加。本文將會(huì)通過采樣和空間設(shè)計(jì)分析強(qiáng)化學(xué)習(xí)在深度自動(dòng)壓縮，同時(shí)提壓縮模型的質(zhì)量。在沒有任何人工操作的情況下，以完全自動(dòng)化的方式獲得了先進(jìn)模型的壓縮結(jié)果。在浮點(diǎn)運(yùn)算縮減為\frac{1}{4}的情況下，實(shí)現(xiàn)了2.8%的精度，高于ImageNet中VGG-16的手動(dòng)壓縮模型。
Abstract
Model-based compression is an effective, facilitating, and expanded model of neural network models with limited computing and low power. However, conventional models of compression techniques utilize crafted features and explore specialized areas for exploration and design of large spaces in terms of size, speed, and accuracy, which usually have returns Less and time is up. This paper will effectively analyze deep auto compression (ADC) and reinforcement learning strength in an effective sample and space design, and improve the compression quality of the model. The results of compression of the advanced model are obtained without any human effort and in a completely automated way. With a 4- fold reduction in FLOP, the accuracy of 2.8% is higher than the manual compression model for VGG-16 in ImageNet.
