We study, for the first time, automated inference on criminality based solely on still face images. Via supervised machine learning, we build four classifiers (logistic regression, KNN, SVM, CNN) using facial images of 1856 real persons controlled for race, gender, age and facial expressions, nearly half of whom were convicted criminals, for discriminating between criminals and non-criminals. All four classifiers perform consistently well and produce evidence for the validity of automated face-induced inference on criminality, despite the historical controversy surrounding the topic …
If given the slightest opportunity, the monkeys will ruin everything:
From the Restoration in 1660, to the end of World War II, the Royal society enforced the scientific method. If you wanted respect and esteem as a scientist, you had to tell us new and interesting things, and you had to show everyone how you knew these new and interesting things from what you saw with your eyes and touched with your hands. […] After World War II, Harvard got the upper hand over the Royal Society, and you no longer have to show your work. Instead, your work must be approved by the most holy synod of mother church – in other words, must pass peer review behind closed doors. Peer Review is new. Attempts to root it in the past of science before World War II are artificial and contrived. Somehow we obtained almost all of science that matters before we had peer review, and since we have had peer review, things have started to go terribly wrong with science. Peer Review is science by social consensus, and Galileo told us that that does not work.
Whatever else time travel may entail, it does not involve changing the past.
— Larry Dwyer (cited here).
Psychology is the canary in the Cathedral.
Bacteria are often invoked as agents of ‘rhizomatic’ (horizontal) disruption of tree-like genetic lineages, so it’s intriguing to see them being proposed (p. 161-2) as cladistic engineers:
If two groups of the same insect ignore each other and only mate with their social circles, they should eventually split into distinct species. These splits occur all the time in nature, and the forces that cause them can take many forms. They could be physical obstacles like mountains or rivers. They could be differences in timing, in the hours or season in which animals are active. They could be incompatible genes that prevent two animals from interbreeding. Anything that stops animals from mating, or that kills or weakens the offspring of those couplings, can create ‘reproductive isolation’ — a gulf that drives species apart. And as [Eugene] Rosenberg had shown, bacteria can cause reproductive isolation, too. By acting as a living barrier that stops two populations from meeting up, microbes could potentially drive the origin of new species. …
Back to basics:
Are there countries with low average scores that tear up the technological track? Mostly not – generally, fairly high average IQ seems to be a prerequisite for creativity in science and mathematics. Necessary, although not sufficient: bad choices (Communism), having the world kick you in the crotch (Mongols), or toxic intellectual fads can all make smart peoples unproductive. […] The exceptions, such as they are, seem to be a result of strong population substructure. India has a low average IQ, but there are distinct subpopulations (castes) that apparently have much higher IQ – although I’d love to see some decent studies on this. With numbers. …
Since the opportunities for XS to agree (in advance) with PZ Myers don’t come along too regularly, it’s worth seizing upon those that do. For anyone who thinks cladistics are important, this point is worth strongly defending:
There are multiple meanings of “fish”. We can use it to refer to specific species or an extant category of animals: salmon are fish, halibut are fish, herring are fish. No one objects to that, and they all understand that if I said “humans are still salmon”, that would be wrong. […] But another way the term is used is as a descriptor for a clade. A taxonomic clade is a “grouping that includes a common ancestor and all the descendants (living and extinct) of that ancestor”. […] So, for instance, humans belong to the mammalian clade, which includes mice and cats and cows. If we have transhuman, part-cyborg descendants, they will still be mammals, because, note, by definition a clade must include all the descendants of an ancestor. We’re trapped! There’s no way our progeny can exit the clade!
In fact, it’s such a sound point, it’s worth generalizing.
(Open thread + links)
The American experiment, conservative insight, and formalism. Confused conservatives. Church and State. Border security (an SF story). Ideological mechanics. Basics of balance. Wrecked names. Responses to the Cowen “neo-reaction” post by Nydwracu, Anti-Gnostic, and VD (with much resonance). The weekly round, plus outliers.
Buried crimes of the Left. Race on campus. The incoherence of Liberal Democracy. America is broken. Defenses of militant atheism, and of mutualism. Jihad denial. Thiel unleashed. Parenthetical remarks (1, 2, 3, 4, 5, 6). Alt-Right rhetoric. Evola and fascism.
Madness in France. Brexit panic spreads (1, 2, 3, 4). The Australian (immigration) model. Swiss still sensible. Venezuela still screwed. State of the jihad, pre-Orlando (1, 2, 3, 4, 5, 6, 7). Islamic ruin in Bangladesh. Oh Canada. The Shenzhen model. Technological de-globalization. “… we are witnessing the end of one era of world history and the dawn of another.”
Trumpenführer panic report. Media fail (1, 2, 3). High-brow Trumpism (an epic troll?). The stupid party. Sad people.
Gawker fun. SV TV. Information suppression.
String-theory by elimination. Evolution of consciousness. Thoughts on the Fermi Paradox. Strange star. The geography of schizophrenia. EQ is BS.
21 Computer goes open-source. Google’s quantum ambitions. Radio disintermediation. An AI movie. China’s deep plans.
The Antikythera Mechanism. Lost cities of Cambodia. Hanson’s AOE book meta-reviews (1, 2, 3, 4, 5).