"Genome-wide analyses of empathy and systemizing: heritability and correlates with sex, education, and psychiatric risk", Warrier et al 2016a:

"Empathy is the drive to identify the mental states of others and respond to these with an appropriate emotion. Systemizing is the drive to analyse or build lawful systems. Difficulties in empathy have been identified in different psychiatric conditions including autism and schizophrenia. In this study, we conducted genome-wide association studies of empathy and systemizing using the Empathy Quotient (EQ) (n = 46,861) and the Systemizing Quotient-Revised (SQ-R) (n = 51,564) in participants from 23andMe, Inc. We confirmed significant sex-differences in performance on both tasks, with a male advantage on the SQ-R and female advantage on the EQ. We found highly significant heritability explained by single nucleotide polymorphisms (SNPs) for both the traits (EQ: 0.11 ± 0.014; P= 1.7x10-14 and SQ-R: 0.12 ± 0.012; P=1.2x10-20) and these were similar for males and females. However, genes with higher expression in the male brain appear to contribute to the male advantage for the SQ-R. Finally, we identified significant genetic correlations between high score for empathy and risk for schizophrenia (P= 2.5x10-5), and correlations between high score for systemizing and higher educational attainment (P= 5x10-4). These results shed light on the genetic contribution to individual differences in empathy and systemizing, two major cognitive functions of the human brain."

Another version of the empathy analysis, using a different measure: a questionnaire of self-rated empathy versus performance on a facial recognition test. Similar but not identical.


"Stereotype (In)Accuracy in Perceptions
of Groups and Individuals", Jussim et al 2015:

"Are stereotypes accurate or inaccurate? We summarize evidence that stereotype accuracy is one of the largest and
most replicable findings in social psychology. We address controversies in this literature, including the long-standing
and continuing but unjustified emphasis on stereotype inaccuracy, how to define and assess stereotype accuracy, and whether stereotypic (vs. individuating) information can be used rationally in person perception. We conclude with
suggestions for building theory and for future directions of stereotype (in)accuracy research."

"Sim-to-Real Robot Learning from Pixels with Progressive Nets", Rusu et al 2016:

"Applying end-to-end learning to solve complex, interactive, pixel-driven control tasks on a robot is an unsolved problem. Deep Reinforcement Learning algorithms are too slow to achieve performance on a real robot, but their potential has been demonstrated in simulated environments. We propose using progressive networks to bridge the reality gap and transfer learned policies from simulation to the real world. The progressive net approach is a general framework that enables reuse of everything from low-level visual features to high-level policies for transfer to new tasks, enabling a compositional, yet simple, approach to building complex skills. We present an early demonstration of this approach with a number of experiments in the domain of robot manipulation that focus on bridging the reality gap. Unlike other proposed approaches, our real-world experiments demonstrate successful task learning from raw visual input on a fully actuated robot manipulator. Moreover, rather than relying on model-based trajectory optimisation, the task learning is accomplished using only deep reinforcement learning and sparse rewards."

A nice use of progressive networks; see also the Schmidhuber paper.

"Safe landing strategies during a fall: Systemic review and meta-analysis", Moon & Sosnoff 2016:

"*Objective*: To systematically synthesize information on safe landing strategies for a fall and quantitatively examine the effects of the strategies to reduce risk of injury from a fall.
Data Sources: PubMed, Web of science, Cumulative Index to Nursing and Allied Health Literature, and Cochrane Library
Study Selection: Databases were searched using the combinations of keywords of “falls”, “strategy”, “impact” and “load”. Randomized control trials, cohort studies, pre-post studies, or cross-sectional studies were included.
Data Extraction: The fall strategies were extracted and categorized by falling direction. Measurements of impact loads that reflect the risk of injuries were extracted (e.g. impact velocity, impact force, fall duration, and impact angle). Hedges g was used as effect size to quantify effect of a protective landing strategy to reduce the impact load.
Data Synthesis: A total of seven landing strategies (squatting, elbow flexion, forward rotation, martial arts rolling, martial arts slapping, relaxed muscle, and stepping) in 13 studies were examined. In general, all strategies, except for the martial arts slapping technique, significantly reduced impact load (g’s=0.73 to 2.70). Squatting was an efficient strategy to reduce impact in backward falling (g=1.77) while elbow flexion with outstretched arms was effective in forward falling (g=0.82). Also, in sideways falling strategies, martial arts rolling (g=2.70) and forward rotation (g=0.82) were the most efficient strategies to reduce impact load.
Conclusions: The result showed that landing strategies have significant effect on reducing impact load during a fall and might be effective to reduce impact load of falling. The current study also highlighted limitations of the previous studies which focused on a young population and self-initiated falls. Further investigation with elderly individuals and unexpected falls is necessary to verify effectiveness and suitableness of the strategies to at-risk population in real-life falls.

A fall is an unexpected event in which an individual comes to rest on the ground floor or lower level 1 . They are one of the leading causes of injury and death among the elderly 2 . An estimated 40% of community-dwelling people aged over 65 years fall at least once a year, and nearly 15% fall twice or more per year 3 . Falls result in 62.5% (2.5 million) of non-fatal injuries of older adults in the United States that require treatment in emergency departments and hospitalization 4 . The direct medical cost for fall related injuries reaches $19 billion annually in the U.S. alone 5 . In addition, as the population ages, the number of annual fall related injuries in the United States is expected to increase to 5.7 million by the year 2030 6 ."

Useful advice for anyone planning on living long enough to become elderly.

Background: http://www.spoon-tamago.com/2016/10/04/animated-gifs-illustrating-the-art-of-japanese-wood-joinery/

"Yahoo's recently open sourced neural network, open_nsfw, is a fine tuned Residual Network which scores images on a scale of 0 to 1 on its suitability for use in the workplace. ...Spurred on by the success above, I explore the possibility of the generation of images for which activations span two different networks. Nguyen et al. has achieved great results on the MIT scene recognition model places-CNN . What happens when we maximize neurons of places-CNN and open_nsfw together?...This program produces the most remarkable results. The images generated range from the garishly explicit to the subtle. But the subtle images are the most fascinating as to my surprise they are only seemingly innocent. These are not adversarial examples per-se. The NSFW elements are all present, just hidden in plain sight. Once you see the true nature of these images, something clicks and it becomes impossible to unsee. I've picked a few of my favorite results for show here."

My sides. Especially the concert, desert, beach, and canyon ones.

"Genome-wide meta-analysis of cognitive empathy: heritability, and correlates with sex, neuropsychiatric conditions and brain anatomy", Warrier et al 2016:

"We conducted a genome-wide meta-analysis of cognitive empathy using the Reading the Mind in the Eyes Test (Eyes Test) in 88,056 Caucasian research participants (44,574 females and 43,482 males) from 23andMe Inc., and an additional 1,497 Caucasian participants (891 females and 606 males) from the Brisbane Longitudinal Twin Study (BLTS). We confirmed a female advantage on the Eyes Test (Cohens d = 0.21, P < 0.001), and identified a locus in 3p26.1 that is associated with scores on the Eyes Test in females (rs7641347, Pmeta = 1.57 x 10-8). Common single nucleotide polymorphisms (SNPs) explained 20% of the twin heritability and 5.6% (± 0.76 ; P = 1.72 x 10-13) of the total trait variance in both sexes. Finally, we identified significant genetic correlation between the Eyes Test and measures of empathy (the Empathy Quotient), openness (NEO-Five Factor Inventory), and different measures of educational attainment and cognitive aptitude, and show that the genetic determinants of striatal volumes (caudate nucleus, putamen, and nucleus accumbens) are positively correlated with the genetic determinants of performance on the Eyes Test."

This is so cyberpunk.

"Do scholars follow Betteridge’s Law? The use of questions in journal article titles", Cook & Plourde 2016 https://www.dropbox.com/s/kootab1g7yr0ecm/2016-cook.pdf

"In journalistic publication, Betteridge’s Law of Headlines stipulates that ‘‘Any headline that ends in a question mark can be answered by the word no.’’ When applied to the titles of academic publication, the assertion is referred to as Hinchcliffe’s Rule and denigrates the use of the question mark in titles as a ‘‘click-bait’’ marketing strategy. We examine the titles of all published articles in the year 2014 from five top-ranked and five mid-range journals in each of six academic fields (n = 7845). We describe the form of questions when they occur, and where a title poses a question that can be answered with a ‘‘yes’’ or ‘‘no’’ we note the article’s substantive answer. We do not find support for the criticism lodged by Betteridge’s Law and Hinchcliffe’s Rule. Although patterns vary by discipline, titles with questions are posed infrequently overall. Further, most titles with questions do not pose yes/no questions. Finally, the few questions that are posed in yes/no terms are actually more often answered with a ‘‘yes’’ than with a ‘‘no.’’ Concerns regarding click-bait questions in academic publications may, therefore, be unwarranted."