These are my links for May 27th through July 28th:
- Statistical Modeling: The Two Cultures – There are two cultures in the use of statistical modeling to reach conclusions from data. One assumes that the data are generated by a given stochastic data model. The other uses algorithmic models and treats the data mechanism as unknown. The statistical community has been committed to the almost exclusive use of data models. This commitment<br />
has led to irrelevant theory, questionable conclusions, and has kept statisticians from working on a large range of interesting current problems. Algorithmic modeling, both in theory and practice, has developed rapidly in fields outside statistics. It can be used both on large complex data sets and as a more accurate and informative alternative to data modeling on smaller data sets. If our goal as a field is to use data to solve problems, then we need to move away from exclusive dependence on data models and adopt a more diverse set of tools.
- Brief history of data visualization – Data visualization is a pretty literal term that means, quite simply, the visual representation of quantitative data. In this course we’ll learn common techniques for visualizing data, as well as some strategies for managing information digitally. But first, a brief history.
- S. Thompson. Motif-index of folk-literature – a classification of narrative elements in folktales, ballads, myths, fables, mediaeval romances, exempla, fabliaux, jest-books, and local legends.
- What is data science? – O’Reilly Radar – We’ve all heard it: according to Hal Varian, statistics is the next sexy job. Five years ago, in What is Web 2.0, Tim O’Reilly said that “data is the next Intel Inside.” But what does that statement mean? Why do we suddenly care about statistics and about data?<br />
<br />
In this post, I examine the many sides of data science — the technologies, the companies and the unique skill sets.
- [1005.0437] A Unifying View of Multiple Kernel Learning – Recent research on multiple kernel learning has lead to a number of approaches for combining kernels in regularized risk minimization. The proposed approaches include different formulations of objectives and varying regularization strategies. In this paper we present a unifying general optimization criterion for multiple kernel learning and show how existing formulations are subsumed as special cases. We also derive the criterion’s dual representation, which is suitable for general smooth optimization algorithms. Finally, we evaluate multiple kernel learning in this framework analytically using a Rademacher complexity bound on the generalization error and empirically in a set of experiments.
These are my links for April 12th through April 22nd:
- Gilbert Harman, Sanjeev Kulkarni – Reliable Reasoning: Induction and Statistical Learning Theory – Reviewed by Kevin Kelly, and Conor Mayo-Wilson, Carnegie Mellon University – Philosophical Reviews – University of Notre Dame – Harman and Kulkarni’s Reliable Reasoning is a welcome attempt to relate machine learning to the philosophy of induction at an introductory level suitable for undergraduates or for professional philosophers and scientists who desire a painless introduction to the subject. In clear, helpful figures and engagingly informal prose, the slender volume summarizes the main results and concepts of statistical learning theory, a particular statistical framework developed by V. N. Vapnik, A. J. Chervonenkis and others for machine learning and other applications (Vapnik and Chervonenkis 1974, Vapnik 1998, 1999, 2000). Harman and Kulkarni also explain how statistical learning theory might shed light on philosophical topics like the problem of induction and the role of simplicity in theory choice. The book is mainly an informal précis of Vapnik (2000), with some important differences.
- Independence and Large Cardinals (Stanford Encyclopedia of Philosophy) – The independence results in arithmetic and set theory led to a proliferation of mathematical systems. One very general way to investigate the space of possible mathematical systems is under the relation of interpretability. Under this relation the space of possible mathematical systems forms an intricate hierarchy of increasingly strong systems. Large cardinal axioms provide a canonical means of climbing this hierarchy and they play a central role in comparing systems from conceptually distinct domains.
- The Sources of Normativity (application/pdf Object) – Whether this is true or not, the moral philosophy of the modern period can be read as a search for the source of normativity. Philosophers in the modern period have come up with four successive answers to the question of what makes morality normative. In brief, they are these: (1) Voluntarism, (2) Realism, (3) Reflective Endorsement, ( 4 ) The Appeal to Autonomy.
- Earliest Known Uses of Some of the Words of Mathematics – These pages attempt to show the first uses of various words used in mathematics. Research for these pages is ongoing, and a citation should not be assumed to be the earliest use unless it is indicated as such.
These are my links for March 23rd through April 12th:
- Earliest Known Uses of Some of the Words of Mathematics – These pages attempt to show the first uses of various words used in mathematics. Research for these pages is ongoing, and a citation should not be assumed to be the earliest use unless it is indicated as such.
- The Illustrated Road to Serfdom – by Friedrich A. Hayek
- cp42252001.pdf (application/pdf Object) – This article discusses the concept of information and its intimate relationship with physics. After an introduction of all the necessary quantum mechanical and information theoretical concepts we analyse Landauer’ s principle which states that the erasure of information is inevitably accompanied by the generation of heat. We employ this principle to rederive a number of results in classical and quantum information theory whose rigorous mathematical derivations are difficult. This demonstrates the usefulness of Landauer’ s principle and provides an introduction to the physical theory of information.
- Data Marketplace : Find, buy and sell data online – Data Marketplace makes it easy for people to find, buy and sell data online.<br />
<br />
Most data must be aggregated, cleaned, and analyzed to extract useful information. It doesn’t make sense that the same person should do all of these things. Data Marketplace connects people who need data with people who are good at collecting, cleaning, and analyzing it.<br />
<br />
People request data that they need. Providers upload data to Data Marketplace, provide descriptive metadata, and set a price. Stored metadata is used to help consumers find relevant data through traditional search engines and when browsing the marketplace.
These are my links for October 25th through October 27th:
- The One Argument Ayn Rand Couldn’t Win — New York Magazine – …as James put it, “a certain insincerity in our philosophic discussions: the potentest of all our premises is never mentioned … What the system pretends to be is a picture of the great universe of God. What it is—and oh so flagrantly!—is the revelation of how intensely odd the personal flavor of some fellow creature is.”No one would have been angrier about this claim, and no one confirms its truth more profoundly, than Ayn Rand. Few fellow creatures have had a more intensely odd personal flavor; her temperament could have neutered an ox at 40 paces. She was proud, grouchy, vindictive, insulting, dismissive, and rash. (One former associate called her “the Evel Knievel of leaping to conclusions.”) But she was also idealistic, yearning, candid, worshipful, precise, and improbably charming. She funneled all of these contradictory elements into Objectivism, the home-brewed philosophy that won her thousands of Cold War–era followers and that seems to be making some noise once again …
- Monsters and the Moral Imagination – The Chronicle Review – The Chronicle of Higher Education – The reasons for this increased monster culture are hard to pin down. Maybe it’s social anxiety in the post-9/11 decade, or the conflict in Iraq—some think there’s an uptick in such fare during wartime. Perhaps it’s the economic downturn. The monster proliferation can be explained, in part, by exploring the meaning of monsters. Popular culture is re-enchanted with meaningful monsters, and even the eggheads are stroking their chins—last month saw the seventh global conference on Monsters and the Monstrous at the University of Oxford.
- The history of management consulting : The New Yorker – … in October of 1910, when Louis Brandeis, a fifty-three-year-old lawyer from Boston, held a meeting at an apartment in New York with a bunch of experts who, at Brandeis’s urging, decided to call what they were experts at “scientific management.” Everyone there—including Frank and Lillian Gilbreth, best known today as the parents in “Cheaper by the Dozen”—had contracted “Tayloritis”: they were enthralled by an industrial engineer from Philadelphia named Frederick Winslow Taylor, who had been ordering people around, scientifically, for years. Speedy Taylor, as he was called, had invented a new way to make money. He would get himself hired by some business; spend a while watching people work, stopwatch and slide rule in hand; write a report telling them how to do their work faster; and then submit an astronomical bill for his services. He is the “Father of Scientific Management”, and, … the grandfather of management consulting.
|
|
