arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2011.12511cs.LGcs.AIcs.DC
收藏

MetaGater:基于联邦元学习的条件信道门控网络快速学习

MetaGater: Fast Learning of Conditional Channel Gated Networks via Federated Meta-Learning

Arizona State University(亚利桑那州立大学);Shanghai Jiao Tong University(上海交通大学)

Sen Lin, Li Yang, Zhezhi He, Deliang Fan, Junshan Zhang

摘要:

虽然深度学习在许多人工智能应用中取得了惊人的成功,但其巨大的模型大小和密集的计算需求对资源有限的节点中的部署提出了严峻的挑战。最近,人们对计算高效的学习方法越来越感兴趣,例如,量化、修剪和信道选通。然而,大多数现有技术不能快速适应不同的任务。在这项工作中,我们提倡一个整体的方法来共同训练的骨干网络和通道门,使动态选择的一个子集的过滤器更有效的本地计算给定的数据输入。特别是,我们开发了一个联邦元学习的方法,共同学习良好的元初始化的骨干网络和门控模块,通过利用不同节点上的学习任务的模型相似性。以这种方式,学习的元门控模块有效地捕获良好元骨干网络的重要滤波器,基于该重要滤波器,可以快速地适配任务特定的条件信道门控网络,即,通过一步梯度下降,从使用该任务的新样本的两阶段过程中的元初始化。联邦元学习算法的收敛性建立在温和的条件下。实验结果证实了我们的方法相比,相关工作的有效性。

英文摘要:

While deep learning has achieved phenomenal successes in many AI applications, its enormous model size and intensive computation requirements pose a formidable challenge to the deployment in resource-limited nodes. There has recently been an increasing interest in computationally-efficient learning methods, e.g., quantization, pruning and channel gating. However, most existing techniques cannot adapt to different tasks quickly. In this work, we advocate a holistic approach to jointly train the backbone network and the channel gating which enables dynamical selection of a subset of filters for more efficient local computation given the data input. Particularly, we develop a federated meta-learning approach to jointly learn good meta-initializations for both backbone networks and gating modules, by making use of the model similarity across learning tasks on different nodes. In this way, the learnt meta-gating module effectively captures the important filters of a good meta-backbone network, based on which a task-specific conditional channel gated network can be quickly adapted, i.e., through one-step gradient descent, from the meta-initializations in a two-stage procedure using new samples of that task. The convergence of the proposed federated meta-learning algorithm is established under mild conditions. Experimental results corroborate the effectiveness of our method in comparison to related work.

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。

↑