Book contents
2 - The landscape of learning machines
Published online by Cambridge University Press: 05 June 2012
Summary
To give an idea of this rapidity, we need only mention that Mr. Babbage believes he can, by his engine, form the product of two numbers, each containing twenty figures, in three minutes. … Now, admitting that such an engine can be constructed, it may be inquired: what will be its utility?
Ada Augusta, Daughter of Lord Byron, Countess of Lovelace (1842)Introduction
In this chapter we outline general themes and ideas related to statistical learning machines. We frame the subject and some of its important analytic processes. Chapter 3 will discuss specific learning machines, with brief discussions of their individual advantages and disadvantages; see Note 1.
Readers wishing to travel a more direct path could consider skipping this chapter, going right to Chapter 3, and then straight on to Chapter 4, where many of the machines are put through their paces on three examples. However, a direct path, free of the context provided by this chapter, may leave the novice disoriented. This is because the geology of machine learning is not a single tectonic plate and is more a vast, still-expanding archipelago, with many separated islands of insight and research, some connected, some not.
Throughout this text we will frequently be quoting, or attempting to translate for the reader, from a wonderful but technically very advanced book, Devroye et al. (1996). This text is magisterial, huge – 635 pages! – dense and highly articulate. Going forward, we will use the abbreviation DGL.
- Type
- Chapter
- Information
- Statistical Learning for Biomedical Data , pp. 14 - 40Publisher: Cambridge University PressPrint publication year: 2011