Multi-Objective Neural Architecture Search for Light-Weight Model | |
Nannan Li2,3; Yaran Chen2,3; Zixiang Ding2,3; Dongbin Zhao2,3; Zhonghua Pang1; Ruisheng Qin1 | |
2019-11 | |
会议日期 | 22-24 November 2019 |
会议地点 | Hangzhou, China |
关键词 | Neural architecture search light-weight multi-objective reinforcement learning image classification |
英文摘要 | Neural architecture search (NAS) has achieved superior performance in visual tasks by automatically designing an effective neural network architecture. In recent years, deep neural networks are increasingly applied to resource-constrained devices. As a result, in addition to the model performance, model size is another very important factor that requires to consider when designing powerful neural network architectures. Therefore, we propose the multi-objective neural architecture search for light-weight model and name it Light-weight NAS. On one hand, the Light-weight NAS introduces Multiply-ACcumulate (MAC) into the optimize objective to get the architecture with fewer parameters. On the other hand, we simplify the search space and adopt weight sharing to make the search process more efficient. Experimental results indicate that the searched architecture can perform competitive classification accuracy with few parameters on the image classification task, while using less computation cost than the most existing multi-objective NAS approaches. |
内容类型 | 会议论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/52189] |
专题 | 复杂系统管理与控制国家重点实验室_深度强化学习 |
作者单位 | 1.Key Laboratory of Fieldbus Technology and Automation of Beijing North China University of Technology 2.School of artificial intelligence, University of Chinese Academy of Sciences 3.State Key Laboratory of Management and Control for Complex Systems, Institute of Automation, Chinese Academy of Sciences |
推荐引用方式 GB/T 7714 | Nannan Li,Yaran Chen,Zixiang Ding,et al. Multi-Objective Neural Architecture Search for Light-Weight Model[C]. 见:. Hangzhou, China. 22-24 November 2019. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论