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 2212.02049cs.LGcs.AIcs.GTmath.OC
收藏

具有成本敏感和策略Agent的分布式随机梯度下降

Distributed Stochastic Gradient Descent with Cost-Sensitive and Strategic Agents

Arizona State University(亚利桑那州立大学)

Abdullah Basar Akbay, Cihan Tepedelenlioglu

摘要:

本研究认为,联邦学习设置成本敏感和战略代理训练的学习模型与服务器。在每一轮中,每个智能体采样一小批训练数据并发送其梯度更新。作为他的小批量大小选择的递增函数,代理招致与数据汇集、梯度计算和通信相关联的成本。代理商可以自由选择他们的小批量,甚至可以选择退出培训。为了降低成本,代理可能会减少他的小批量大小,这也可能导致梯度更新的噪声水平增加。服务器可以提供奖励来补偿代理的成本并激励他们的参与,但她缺乏验证代理的真实小批量大小的能力。为了解决这一挑战,建议的奖励机制评估每个代理的梯度的质量,根据其距离的参考,这是从其他代理提供的梯度构造。结果表明,建议的奖励机制有一个合作的纳什均衡,其中代理商确定的小批量大小的选择,根据服务器的请求。

英文摘要:

This study considers a federated learning setup where cost-sensitive and strategic agents train a learning model with a server. During each round, each agent samples a minibatch of training data and sends his gradient update. As an increasing function of his minibatch size choice, the agent incurs a cost associated with the data collection, gradient computation and communication. The agents have the freedom to choose their minibatch size and may even opt out from training. To reduce his cost, an agent may diminish his minibatch size, which may also cause an increase in the noise level of the gradient update. The server can offer rewards to compensate the agents for their costs and to incentivize their participation but she lacks the capability of validating the true minibatch sizes of the agents. To tackle this challenge, the proposed reward mechanism evaluates the quality of each agent's gradient according to the its distance to a reference which is constructed from the gradients provided by other agents. It is shown that the proposed reward mechanism has a cooperative Nash equilibrium in which the agents determine the minibatch size choices according to the requests of the server.

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。

↑