Publications by matloff
New Book on Machine Learning
I’m nearing completion of writing my new book, The Art of Machine Learning: Algorithms+Data+R, to be published by the whimsically named No Starch Press. I’m making a rough, partial draft available, and welcome corrections, suggestions and comments. I’ve been considering doing such a project for some time, intending to write a book that woul...
2011 sym
New Book on Machine Learning
I’m nearing completion of writing my new book, The Art of Machine Learning: Algorithms+Data+R, to be published by the whimsically named No Starch Press. I’m making a rough, partial draft available, and welcome corrections, suggestions and comments. I’ve been considering doing such a project for some time, intending to write a book that woul...
2011 sym
Efron Updates breiman’s “two cultures” essay
The June 2020 issue of JASA features a highly insightful essay by Brad Efron, dean of the world’s statisticians. The article is accompanied by commentary by a number of statistical luminaries. Important, indeed central questions are raised. I would hope JASA runs more pieces of this nature. The essay may be viewed as an update of the classic 2...
9798 sym
Efron Updates breiman’s “two cultures” essay
The June 2020 issue of JASA features a highly insightful essay by Brad Efron, dean of the world’s statisticians. The article is accompanied by commentary by a number of statistical luminaries. Important, indeed central questions are raised. I would hope JASA runs more pieces of this nature. The essay may be viewed as an update of the classic 2...
9798 sym
The Notion of “Double Descent”
I tend to be blase’ about breathless claims of “new” methods and concepts in statistics and machine learning. Most are “variations on a theme.” However, the notion of double descent, which has come into prominence in the last few years, is something I regard as genuinely new and very relevant, shedding important light on the central iss...
6761 sym R (676 sym/1 pcs) 2 img
The Notion of “Double Descent”
I tend to be blase’ about breathless claims of “new” methods and concepts in statistics and machine learning. Most are “variations on a theme.” However, the notion of double descent, which has come into prominence in the last few years, is something I regard as genuinely new and very relevant, shedding important light on the central iss...
6390 sym R (655 sym/1 pcs) 2 img
Issues in Differential Privacy
Differential privacy (DP) is an approach to maintaining the privacy of individual records in a database, while still allowing statistical analysis. It is now perceived as the go-to method in the data privacy area, enjoying adoption by the US Census Bureau and several major firms in industry, as well as a highly visible media presence. DP has de...
13885 sym
What’s Not What in Differential Privacy
I take my title here from the “too clever by half” paper, “What’s Not What with Statistics” of many years ago. Or I just as appropriately could have borrowed the old Leo Breiman classic title “A Tale of Two Cultures,” comparing the statistics and computer science (CS) communities. Differential privacy (DP) is an approach to maintain...
14647 sym