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 2110.04124cs.LGeess.IVeess.SP
收藏

嵌入式神经表示网络

Ensemble Neural Representation Networks

Iran University of Science and Technology(伊朗科技大学);University of Tehran(德黑兰大学);McGill University(麦吉尔大学);Lund University(隆德大学);Sharif University of Technology(谢里夫理工大学)

Milad Soltany Kadarvish, Hesam Mojtahedi, Hossein Entezari Zarch, Amirhossein Kazerouni, Alireza Morsali, Azra Abtahi, Farokh Marvasti

摘要:

内隐神经表征(INR)是一种以连续格式纸存储各种信号的方法,近年来引起了广泛的关注。现有的INR网络需要冗长的训练过程和高性能的计算资源。在本文中,我们提出了一种新的次优集成架构INR,解决了上述问题。在该体系结构中,表示任务被划分为多个子任务,由独立的子网络完成。我们表明,建议的合奏INR架构的性能可能会下降,如果子网络的维度增加。因此,提出一种优化算法来找到次优的集成网络结构是至关重要的,这是本文所做的。根据仿真结果,所提出的架构不仅具有显着较少的浮点运算(FLOPs)和更少的训练时间,但它也有更好的性能,在峰值信噪比(PSNR)相比,其同行。

英文摘要:

Implicit Neural Representation (INR) has recently attracted considerable attention for storing various types of signals in continuous forms. The existing INR networks require lengthy training processes and high-performance computational resources. In this paper, we propose a novel sub-optimal ensemble architecture for INR that resolves the aforementioned problems. In this architecture, the representation task is divided into several sub-tasks done by independent sub-networks. We show that the performance of the proposed ensemble INR architecture may decrease if the dimensions of sub-networks increase. Hence, it is vital to suggest an optimization algorithm to find the sub-optimal structure of the ensemble network, which is done in this paper. According to the simulation results, the proposed architecture not only has significantly fewer floating-point operations (FLOPs) and less training time, but it also has better performance in terms of Peak Signal to Noise Ratio (PSNR) compared to those of its counterparts.

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。

↑