Machine Learning Foundations –lesson 1
Key Essence of Machine Learning
1.exists some ‘underlying pattern’ to be learned
2.but no programmable (easy) definition
3.somehow there is data about the pattern
第二个题目是可以用计算机程序解决的
第一题小孩哭这个事情没有规律可循
Formalize the Learning Problem
1.input: x ∈ X (customer application): 模型的输入
2output: y ∈ Y (good/bad after approving credit card)–模型的输出
3.unknown pattern to be learned ⇔ target function:–能够完美拟合数据的函数
f : X → Y (ideal credit approval formula)
4.data ⇔ training examples: D = {(x1, y1),(x2, y2), · · · ,(xN, yN)}—样本
(historical records in bank)
5.hypothesis ⇔ skill with hopefully good performance:—-学习到模型,假设函数
g : X → Y (‘learned’ formula to be used)
Machine Learning and Data Mining and Artfical Intelligence
数据挖掘与机器学习他们之间的接线很模糊
机器学习与人工只能
例如很热的强化学习,纯粹的机器学习是基于统计的,还有基于神经网络的深度学习都可以解决AI(自动驾驶,智能语音,对话机器人等)