图 书 馆
快餐：对数线性时间的近似核膨胀Fast Food: Approximating Kernel Expansion in Loglinear Time |
|
| 课程网址: | http://videolectures.net/nipsworkshops2012_smola_kernel/ |
| 主讲教师: | Alex Smola |
| 开课单位: | 亚马逊公司 |
| 开课时间: | 2013-01-18 |
| 课程语种: | 英语 |
| 中文简介: |  快速评估非线性函数类别的能力对于非参数估计至关重要。我们提出了一种对随机厨房水槽的改进，该水槽可为d维中的n个基函数提供O（n log d）计算和O（n）存储，而不会牺牲精度。我们展示了如何仅通过更改投影矩阵的光谱分布即可调整内核的正则化属性。实验表明，我们达到了与完整内核扩展相同的精度，随机厨房接收器的速度提高了100倍，内存减少了1000倍。 |
| 课程简介: | The ability to evaluate nonlinear function classes rapidly is crucial for nonparametric estimation. We propose an improvement to random kitchen sinks that offers O(n log d) computation and O(n) storage for n basis functions in d dimensions without sacrificing accuracy. We show how one may adjust the regularization properties of the kernel simply by changing the spectral distribution of the projection matrix. Experiments show that we achieve identical accuracy to full kernel expansions and random kitchen sinks 100x faster with 1000x less memory. |
| 关 键 词: | 非线性函数; 内核扩展 |
| 课程来源: | 视频讲座网 |
| 数据采集: | 2020-11-05：zyk |
| 最后编审: | 2020-11-05：zyk |
| 阅读次数: | 46 |