机器学习的高斯过程
Gaussian processes for machine learning
Carl Edward Rasmussen · 初版 2005 年
5人推荐
6位读者提及
6次提及
Gaussian processes（高斯过程，简称 GPs）提供了一种核机器学习方法。本书全面系统地探讨了 GPs 在机器学习中的理论与实践问题。内容详尽透彻、自成体系，主要面向机器学习和应用统计领域的研究人员与学生。（摘自本书网站：http://www.gaussianprocess.org/gpml/ ）
Hacker News 上怎么说
正在加载完整评论……
……还有 sklearn.gaussian_process [1]，它提供了高斯过程分类与回归，属于贝叶斯学习算法，可以看作是支持向量机（SVM）的类似物。拉斯穆森的《Gaussian Processes for Machine Learning》（2006）是一本极佳的入门书，尽管成书稍微有点早了。[2]。[0] https://scikit-learn.org/stable/modules/mixture.html [1] https://scikit-learn.org/stable/modules/gaussian_process.htm... [2] http://gaussianprocess.org/gpml/chapters/
…also sklearn.gaussian_process [1] which offers Gaussian process classification and regression, Bayesian learning algorithms which can be thought of as analogs of the support vector machine. Rasmussen’s Gaussian Processes for Machine Learning (2006) is a great introduction despite being a little old. [2]. [0] https://scikit-learn.org/stable/modules/mixture.html [1] https://scikit-learn.org/stable/modules/gaussian_process.htm... [2] http://gaussianprocess.org/gpml/chapters/
……计算核函数和均值预测时带来的尖峰。这是一个分两步的过程。要理解确切的公式，我强烈推荐文章中提到的资源 [2]——《Gaussian Processes for Machine Learning》第二章。如果想看教学视频，我推荐 [3]，来自康奈尔大学的一门 ML 课程。至于如何高效实现高斯过程（GPs），这依然是个挑战：numpy 虽然有一堆矩阵-向量函数……
…spikes that came with computing the kernel and mean predictions. It was a two-fold process. For understanding the exact equations, I would highly recommend the resource mentioned in the article [2], Gaussian Processes for Machine Learning Chapter 2. For a lecture video, I would recommend [3] which is from a ML course at Cornell. As for implementing GPs efficiently, this is still a challenge: numpy has a bunch of matrix-vector functions…
你看过《Gaussian Processes for Machine Learning》[0] 吗？全书都可以在上述 URL 免费在线阅读。[0] http://www.gaussianprocess.org/gpml/
Have you seen Gaussian Processes for Machine Learning [0]? The entire text is freely available online at the mentioned URL. [0] http://www.gaussianprocess.org/gpml/
……Chebira http://www.e-booksdirectory.com/details.php?ebook=2852 《How Are We To Know?》，尼尔斯·J·尼尔森著 http://www.e-booksdirectory.com/details.php?ebook=2710 《强化学习导论》（Reinforcement Learning: An Introduction） http://www.e-booksdirectory.com/details.php?ebook=1825 《Gaussian Processes for Machine Learning》 http://www.e-booksdirectory.com/details.php?ebook=1774 《Machine Learning, Neural and Statistical Classification》 http://www.e-booksdirectory.com/details.php?ebook=1118 《Introduction To Machine Learning》 http://www.e-booksdirectory.com/details.php?ebook=1117……
…Chebira http://www.e-booksdirectory.com/details.php?ebook=2852 How Are We To Know? by Nils J. Nilsson http://www.e-booksdirectory.com/details.php?ebook=2710 Reinforcement Learning: An Introduction http://www.e-booksdirectory.com/details.php?ebook=1825 Gaussian Processes for Machine Learning http://www.e-booksdirectory.com/details.php?ebook=1774 Machine Learning, Neural and Statistical Classification http://www.e-booksdirectory.com/details.php?ebook=1118 Introduction To Machine Learning http://www.e-booksdirectory.com/details.php?ebook=1117…
你可能对《Gaussian Processes for Machine Learning》这本书感兴趣：http://gaussianprocess.org/gpml/（免费提供）
You may be interested in the book Gaussian Processes for Machine Learning: http://gaussianprocess.org/gpml/ (freely available)
……免费与便宜之间的重要区别。获取门槛变低会带来各种催化效应。例如，免费教材可以随时被引用：“关于此问题的更详细讨论，请参阅《Gaussian Processes for Machine Learning》第四章。”
…important difference between free & cheap. Easier access has all sorts of catalyzing effects. For example, a free textbook can be referred to. "For a more detailed discussion of this read chapter 4 of Gaussian Processes for Machine Learning."