Archive for data privacy

weapons of math destruction [book review]

Posted in Books, Kids, pictures, Statistics, University life with tags , , , , , , , , , , , , , , , on December 15, 2016 by xi'an

wmd As I had read many comments and reviews about this book, including one by Arthur Charpentier, on Freakonometrics, I eventually decided to buy it from my Amazon Associate savings (!). With a strong a priori bias, I am afraid, gathered from reading some excerpts, comments, and the overall advertising about it. And also because the book reminded me of another quantic swan. Not to mention the title. After reading it, I am afraid I cannot tell my ascertainment has changed much.

“Models are opinions embedded in mathematics.” (p.21)

The core message of this book is that the use of algorithms and AI methods to evaluate and rank people is unsatisfactory and unfair. From predicting recidivism to fire high school teachers, from rejecting loan applications to enticing the most challenged categories to enlist for for-profit colleges. Which is indeed unsatisfactory and unfair. Just like using the h index and citation ranking for promotion or hiring. (The book mentions the controversial hiring of many adjunct faculty by KAU to boost its ranking.) But this conclusion is not enough of an argument to write a whole book. Or even to blame mathematics for the unfairness: as far as I can tell, mathematics has nothing to do with unfairness. Some analysts crunch numbers, produce a score, and then managers make poor decisions. The use of mathematics throughout the book is thus completely inappropriate, when the author means statistics, machine learning, data mining, predictive algorithms, neural networks, &tc. (OK, there is a small section on Operations Research on p.127, but I figure deep learning can bypass the maths.) Continue reading

AISTATS 2016 [#2]

Posted in Kids, pictures, Running, Statistics, Travel, University life, Wines with tags , , , , , , , , , , , , , , , on May 13, 2016 by xi'an

The second and third days of AISTATS 2016 passed like a blur, with not even the opportunity to write my impressions in real time! Maybe long tapa breaks are mostly to blame for this… In any case, we had two further exciting plenary talks about privacy-preserving data analysis by Kamalika Chaudhuri and crowdsourcing and machine learning by Adam Tauman Kalai. The talk by Kamalika was covering recent results by Kamalika and coauthors about optimal privacy preservation in classification and a generalisation to correlated data, with the neat notion of a Markov Quilt.  Other talks that same day also dwelt on this privacy issue, but I could not be . The talk by Adam was full of fun illustrations on humans training learning systems (with the unsolved difficulty of those humans deliberately mis-training the system, as exhibited recently by the short-lived Microsoft Tay experiment).

Both poster sessions were equally exciting, with the addition of MLSS student posters on the final day. Among many, I particularly enjoyed Iain Murray’s pseudo-marginal slice sampling, David Duvenaud’s fairly intriguing use of early stopping for non-parametric inference,  Garrett Bernstein’s work on aggregated Markov chains, Ye Wang’s scalable geometric density estimation [with a special bonus for his typo on the University of Turing, instead of Torino], Gemma Moran’s and Chengtao Li’s posters on determinantal processes, and Matej Balog’s Mondrian forests with a Laplace kernel [envisioning potential applications for ABC]. Again, just to mention a few…

The participants [incl. myself] also took one evening off to visit a sherry winery in Jerez, with a well-practiced spiel on the story of the company, with some building designed by Gutave Eiffel, and with a wine-tasting session. As I personally find this type of brandy too strong in alcohol, I am not a big fan of sherry but it was nonetheless an amusing trip! With no visible after-effects the next morning, since the audience was as large as usual for Adam’s talk [although I did not cross a machine-learning soul on my 6am run…]

In short, I enjoyed very much AISTATS 2016 and remain deeply impressed by the efficiency of the selection process and the amount of involvement of the actors of this selection, as mentioned earlier on the ‘Og. Kudos!


Posted in Books, Statistics with tags , , , , , , , , on December 1, 2015 by xi'an

WariseWhile in Warwick this week, I borrowed a recent issue (Oct. 08, 2015) of Nature from Tom Nichols and read it over diners in a maths house. Its featured topic was reproducibility, with a long initial (or introductory) article about “Fooling ourselves”, starting with an illustration from Andrew himself who had gotten a sign wrong in one of those election studies that are the basis of Red State, Blue State. While this article is not bringing radically new perspectives on the topic, there is nothing shocking about it and it even goes on mentioning Peter Green and his Royal Statistical Society President’s tribune about the Sally Clark case and Eric-Jan Wagenmakers with a collaboration with competing teams that sounded like “putting one’s head on a guillotine”. Which relates to a following “comment” on crowdsourcing research or data analysis.

I however got most interested by another comment by MacCoun and Perlmutter, where they advocate a systematic blinding of data to avoid conscious or unconscious biases. While I deem the idea quite interesting and connected with anonymisation techniques in data privacy, I find the presentation rather naïve in its goals (from a statistical perspective). Indeed, if we consider data produced by a scientific experiment towards the validation or invalidation of a scientific hypothesis, it usually stands on its own, with no other experiment of a similar kind to refer to. Add too much noise and only noise remains. Add too little and the original data remains visible. This means it is quite difficult to calibrate the blinding mechanisms in order for the blinded data to remain realistic enough to be analysed. Or to be different enough from the original data for different conclusions to be drawn. The authors suggest blinding being done by a software, by adding noise, bias, label switching, &tc. But I do not think this blinding can be done blindly, i.e., without a clear idea of what the possible models are, so that the perturbed datasets created out of the original data favour more one of the models under comparison. And are realistic for at least one of those models. Thus, some preliminary analysis of the original or of some pseudo-data from each of the proposed models is somewhat unavoidable to calibrate the blinding machinery towards realistic values. If designing a new model is part of the inferential goals, this may prove impossible… Again, I think having several analyses run in parallel with several perturbed datasets quite a good idea to detect the impact of some prior assumptions. But this requires statistically savvy programmers. And possibly informative prior distributions.