This post is on the heels of a paper1 we recently published (with Éva Czabarka) on the degeneracy problem of the Maximum Entropy Method. The post is fairly long as it also provides a tutorial on the topic.
When we are trying to understand a system or predict its behavior, we usually do that based on limited information. In everyday situations we make our decisions and choices using some sort of “inner” probability distribution shaped by past experience. However, human decision-making tends to be biased; just this wiki page enlists over 150 such biases. Naturally, this raises the question whether there is a way to make in-silico, unbiased quantitative inferences based on limited information.
Edwin T. Jaynes, 1922-98
(Washington U. Libraries)
The father of a principled and quantitative approach to unbiased inference making is physicist Edwin T. Jaynes. Starting off his career under the tutelage of none other than Eugene (Jenő) Wigner, he brought significant contributions in several areas of physics, most notably in statistical mechanics and its applications. Standing on the shoulders of Laplace, Bayes and Shannon, Jaynes started a revolution in statistical inference with his two seminal papers2,3 from 1957. He based his inference method on the notion of statistical ensembles, much in the same way that statistical mechanics uses the ensembles of microstates to describe macroscopic properties of matter.
In his two celebrated papers Jaynes has re-derived many results from both equilibrium statistical mechanics and the time-dependent density matrix formalism of quantum physics using only the Maximum Entropy Principle. The plot below shows what it means to write papers for the ages: his two papers from 1957 have been accumulating citations exponentially, with citations doubling roughly every ten years.
Citations to Jaynes’s 1957 papers over the years.
Source: Web of Science; Thomson-Reuters.
(Update added April 20, 2015, see the end of this post.)
I have been procrastinating for while about setting up this blog, but recent events have pushed me over the bump, so here it is. In the past week about 145 editors have resigned (including yours truly) from the editorial membership board of the journal Scientific Reports, for a cause supported now by over 550 signed supporters and counting. The editorial membership board of SciRep has thousands of members. Sorry for the long post, but the issues are serious.
On March 24, 2015, Scientific Reports (SciRep from here on), an open access journal owned by NPG/Palgrave Macmillan has sent out a letter to its editorial board members, alerting us about an experiment in peer review they were conducting. See here for additional info. In a nutshell, SciRep has launched a small-scale experiment, only in biology, which would secure fast peer review for those authors who are willing to pay for it. It guarantees to return the reviews to the authors within three weeks, or their money back. To accomplish this, SciRep employs the services of a 3rd party, Research Square, using their peer review system Rubriq. More details are revealed here. Currently, fast-track service seems to be set at \$750 a pop which would be in addition to the \$1,495 fee for publishing the paper. It is not clear how much of that goes to the reviewers, I suspect a small fraction (\$100 ?) as the rest goes to the for-profit parts of the setup.
This has led to an outcry from the part of the editors, for reasons outlined in an open letter that can be found here: http://www.peerreviewneutrality.org/ . This letter is also a letter of resignation by those editors, effective immediately. If you are an editor for this journal, please read this letter and consider the reasons for it and the arguments below; hopefully you’ll sign it too. Even if you are not an editor, as a scientist, please consider supporting the cause. As I will try to outline below, the issues are larger than this particular problem, and we all need to be aware of them.