记忆有界稀疏线性回归的最小最大速率Minimax rates for memory-bounded sparse linear regression |
|
课程网址: | http://videolectures.net/colt2015_steinhardt_linear_regression/ |
主讲教师: | Jacob Steinhardt |
开课单位: | 斯坦福大学 |
开课时间: | 2015-08-20 |
课程语种: | 英语 |
中文简介: | 我们建立了在给定的比特存储边界中估计参数所需的样本数量的最小-最大下界,其中参数误差。当回归器的协方差是单位矩阵时,我们还提供了一种使用样本来实现误差的算法。我们的下界也适用于更通用的通信有界设置,其中,允许关于每个样本(自适应地)传递最多比特的信息,而不是内存有界。 |
课程简介: | We establish a minimax lower bound of on the number of samples needed to estimate the parameters in a given a memory bound of bits, where parameter error. When the covariance of the regressors is the identity matrix, we also provide an algorithm that uses samples to achieve error . Our lower bound also holds in the more general communication-bounded setting, where instead of a memory bound, at most bits of information are allowed to be (adaptively) communicated about each sample. |
关 键 词: | 内存边界; 估计维度; 内存限制 |
课程来源: | 视频讲座网 |
数据采集: | 2022-12-08:chenjy |
最后编审: | 2023-03-16:liyy |
阅读次数: | 20 |