arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1904.06307cs.LGstat.ML
收藏

重温Lmser及其基于卷积层的进一步发展

Revisit Lmser and its further development based on convolutional layers

Shanghai Jiao Tong University(上海交通大学);The Chinese University of Hong Kong(香港中文大学)

Wenjing Huang, Shikui Tu, Lei Xu

摘要:

自组织网络的最小均方误差重构算法(Least Mean Square Error Reconstruction for Self-Organizing Network,简称Lmser)于1991年提出,是对传统自编码器(AE)的进一步发展,它将传统AE的结构相对于中心编码层进行折叠,从而产生对称权值和神经元的特征,以及联合监督学习和无监督学习的特征。然而,由于当时缺乏计算资源和大数据,其优势仅在一个隐藏层实现中得以展示。本文从深度学习的角度重新审视Lmser,开发了基于多卷积层的Lmser网络,更适合于图像相关的任务,并通过对图像识别、重建、关联回忆等方面的初步演示,确认了几个Lmser函数。实验证明,Lmser确实如原文所述,并且它在各种应用中具有有希望的性能。

英文摘要:

Proposed in 1991, Least Mean Square Error Reconstruction for self-organizing network, shortly Lmser, was a further development of the traditional auto-encoder (AE) by folding the architecture with respect to the central coding layer and thus leading to the features of symmetric weights and neurons, as well as jointly supervised and unsupervised learning. However, its advantages were only demonstrated in a one-hidden-layer implementation due to the lack of computing resources and big data at that time. In this paper, we revisit Lmser from the perspective of deep learning, develop Lmser network based on multiple convolutional layers, which is more suitable for image-related tasks, and confirm several Lmser functions with preliminary demonstrations on image recognition, reconstruction, association recall, and so on. Experiments demonstrate that Lmser indeed works as indicated in the original paper, and it has promising performance in various applications.

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。

↑