Machine learning: Trends,perspectives, and prospects读书笔记

M. I. Jordan和T. M. Mitchell在Science上发表了一篇论文Machine learning: Trends,perspectives, and prospects，这篇论文对机器学习的基本概念，发展状况，应用，常见的学习算法做了介绍。对于刚入门来说，构建一个比较基本的机器学习的概念图。

在人工智能里，机器学习作为一种解决计算机视觉，语音识别，自然语言处理，机器控制等应用的可行的软件实现方法。

 

机器学习需要解决的两大核心问题：

Machine learning is a discipline focused on two interrelated questions: How can one construct computer systems that automatically improve through experience and What are the fundamental statistical-computational-information-theoretic laws that govern all learning systems, including computers, humans, and organizations? 

1.如何构建一个可以通过学习经验而不断完善的计算机系统

2.归纳出可计算的涵盖所有学习系统的规律

 

对于一个学习算法，需要考虑一下问题：

Whatever the learning algorithm, a key scientific and practical goal is to theoretically characterize the capabilities of specific learning algorithms and the inherent difficulty of any given learning problem: How accurately can the algorithm learn from a particular type and volume of training data? How robust is the algorithm to errors in its modeling assumptions or to errors in the training data? Given a learning problem with a given volume of training data, is it possible to design a successful algorithm or is this learning problem fundamentally intractable?Given a learning problem with a given volume of training data, is it possible to design a successful algorithm or is this learning problem
fundamentally intractable?

1.该算法的适用范围

2.学习算法固有的局限性：

（1）该算法如何从特定的类型和一定数量的训练数据精确地学习呢？

（