Book Review: Machine Learning for Hackers by Conway and White

Machine Learning for Hackers by Drew Conway
My rating: 5 of 5 stars

Machine Learning for Hackers is not a reference book or a standard programming tutorial on machine learning. For references, you go to Hastie, Tibshirani,and Friedman's 'The Elements of Statistical Learning.' For tutorials, there are a fair number of sources that could walk you through the use of regression, data exploration, classifiers, principle component analysis, etc functions in R. But what MLfH gives you are Drew Conway and John Myles White. And they don't teach skills, they are passing on wisdom of how to work with data, how data needs to be explored, understood, manipulated, and finally, using machine learning methods to gain understanding.

In computer modeling in general and data analysis in particular, one thing that is often hard to convey is that the purpose of computing is not numbers, but insight. The effects of this problem is seen in graduate from even the best schools knowing how to drive a computer program, but not knowing how to interpret results or how to ask a question, then taking the results from that and asking the next question. The course we teach and the texts that we use do not help. Our courses are each siloed to present a distinct portion of the total body of knowledge. Textbooks are often either theoretical or intended to provide a glimpse of application, but always in bounded chunks. Computer application books are often built around the capabilities of program in question, but often stop at the edge of the capabilities of the application or environment in question. What is needed is not to tech methods or tools, but to teach wisdom. The ideal is to be able to sit side by side with an expert who can walk through a data set and ask questions, get answers, and to think about what to do next, whether the answer is what was expected or not.

This is what Conway and White do. For each topic, they open up with a discussion of the problem type and the tools, and sometimes with a toy example. But then they go through a substantive example. And the narrative text is where they shine. They take a messy dataset (often the publicly available/accessible form) and work through what needs to be massaged to get it into useable form. Next is processing the data into the R data type needed for the analysis. Then initial exploratory steps where you gain understanding of problem, and how to analyze it, finally analysis and presentation.

I've been taught that in learning a programming language, it is often beneficial to have two books for reference (other than tutorials), one that is a proper reference (i.e. how to do something), and one a morality reference, how you should approach doing something. In data analysis, you should know the theory/methodology and how to use the tools at hand to apply the methodology, but also how to think about problems. And short of an apprenticeship with a master, MLfH does very well in this.

Disclaimer: I received a free electronic copy of this book from the OReilly Blogger Program. More information on this book can be found at the book web site.

Pittsburgh Symphony: The rhythm of modern life

[Original at the Pittsburgh Symphony Blogs]
This weekend Maestro Previn returned again to conduct the Pittsburgh Symphony Orchestra, and this time to conduct the premier of his own work. As a taste, we were treated to the Brass Quartet from Carnegie Mellon University performing some of Previn's other work.
Listening to the opening of Previn's Triple Concerto for Horn, Trumpet, Tuba and Orchestra immediately following Haydn's Symphony No. 102, I was struck by a contract in feel. Haydn's Symphony sounded smooth, peaceful, and gentle; almost bucolic. And compared to that the Triple Concerto was full of rythmic brasses calling for attention along with a series of fanfares.
In many ways much like the many things that call for our attention in our modern, urban life. It had the same effect of the opening of American in Paris. I had in my mind the thought of a young person starting out in the world, with many things to call for his attention, time, and energy. The ups and downs of young adult life. And this was brought to a close with a rapid movement through the end, and then the piece was done. But it seemed to be a story in progress, because whatever was ahead for our friend, it is never the end of the story, only one more milestone of this thing called life, however momentous it may seem at the moment. And while the piece had an end, real life always has a continuation.
There are many commentators on how life has changed within memory. The speed of change, the need to adapt on an ongoing basis, and for the young, the tendency to treat immediate events out of proportion to their place in an ever-changing and on-going world.

Pittsburgh Symphony: Thinking of Classics and Enigmas

[Original post is at the Pittsburgh Symphony Blog site] Discussing Enigmas - A post-concert discussion of Elgar Enigma Variations, Pittsburgh Symphony Orchestra Overlook Room

This past Sunday’s concert included a workshop and post-concert discussion led by teaching artist Christina Farrell. For Elgar’s Enigma Variations, the easy choice of topics would be to look at three enigmas directly presented by Elgar. But as we talked, conversation covered another topic, does the direction or intent of the creator of a piece play into the performance of the piece? And from there, what makes a creative work worthy of being one of the classics?

To be sure, the use of hidden themes and puzzles within musical works has been done before and since. Several composers have pieces full of references to friends, acquaintances, and contemporaries. But this is par for the course in the classics. My high school English teacher was fond of saying that the Greek classics include everything. Participants in one web site that focuses on identifying themes in movies and TV shows delight in noting that certain themes have very old origins. But this does not lessen the treatment of these themes or use of these tropes. There must be another criteria.

I am involved in a creative field. And in my field, some works are known and viewed as seminal works, definitive in their topic over anything before or since and viewed as original contributions of high degree and quality. Yet even these explicitly reference the works of others, not all of which are in themselves worthy of the same acclaim. The creators and all those who are qualified to review the work will acknowledge it openly, but something separates these from others of their kind.

Mortimer Alder once wrote an introduction to one collection of Great Books. In it he describes three criteria that he and his fellow collaborators used to choose and cull that collection. One was the work’s contemporary significance (which implied that it was timeless beyond the period of original creation); second was its re-readability, or the fact that the work could be read again and again, with deeper understanding every time; third was its relevance to a number of great ideas that occupied a number of great thinkers before and since, the participation in the work in the great conversation through the author’s reading of related works of before, and the development of ideas that have inspired reaction from those who follow.

And here in music, it is not just the theme or subject of the work that makes it meet similar criteria. The price of admission into a classic worth listening repeatedly is the quality of the music. But if it was only quality we cared about, we may stop listening to new music, because there is more than enough of the traditional classics of quality to keep us wonderfully engaged. But to add to what has come before a work must also have something to say. A response to what has come before, and something that is worth responding to.

Parenting month 16: Toddler in motion

Oops, missed a month. These past two months have been marked by T moving a lot more. He goes up and down stairs. He stands up and walks around. And he can climb on chairs and couches. And go exploring cabinets and anything else. Definitely have to work to keep track of him now.

Another major milestone is that mommy finally took a trip away without T. So for four nights, mommy got to sleep :-) And when she came back baby, daddy, and house were all intact and reasonably well fed.

T is still somewhat lacking in the socialization department. He is the one who is playing alone when everyone else is crowded in one corner at day care. On the other hand, he is ok being in places with lots of other people, he will just ignore them.

On the developmental front, one comment that we have gotten is that he has a reasonably good attention span (for a 16 month old). He will sit through reading books (and we have him trained to take care of turning the pages). And he will play with one thing for a reasonable amount of time.

New things to look forward too. The grandparents are coming, taking shifts over the next 6-8 months. Lots of fun for T. Exercise for grandparents. And mommy and daddy are looking forward to having fun leaving T at home :-)