Toggle light / dark theme

Researchers at the Allen Institute for Artificial Intelligence (AI2) have created a machine learning algorithm that can produce images using only text captions as its guide. The results are somewhat terrifying… but if you can look past the nightmare fuel, this creation represents an important step forward in the study of AI and imaging.

Unlike some of the genuinely mind-blowing machine learning algorithms we’ve shared in the past—see here, here, and here —this creation is more of a proof-of-concept experiment. The idea was to take a well-established computer vision model that can caption photos based on what it “sees” in the image, and reverse it: producing an AI that can generate images from captions, instead of the other way around.

This is a fascinating area of study and, as MIT Technology Review points out, it shows in real terms how limited these computer vision algorithms really are. While even a small child can do both of these things readily—describe an image in words, or conjure a mental picture of an image based on those words—when the Allen Institute researchers tried to generate a photo from a text caption using a model called LXMERT, it generated nonsense in return.

Comet 67P/C-G is a dusty object. As it neared its closest approach to the Sun in late July and August 2015, instruments on Rosetta recorded a huge amount of dust enshrouding the comet.

This is tied to the comet’s proximity to our parent star, its heat causing the comet’s nucleus to release gases into space, lifting the dust along. Spectacular jets were also observed, blasting more dust away from the comet. This disturbed, ejected material forms the ‘coma’, the gaseous envelope encasing the comet’s nucleus, and can create a beautiful and distinctive tail.

A single image from Rosetta’s OSIRIS instrument can contain hundreds of dust particles and grains surrounding the 4 km-wide comet nucleus. Sometimes, even larger chunks of material left the surface of 67P/C-G — as shown here.

SpaceX’s Starlink has showed its utility in connecting far-flung locations to the internet quickly and relatively simply in Washington, where like much of the west coast wildfires have caused enormous damage to rural areas. A couple small towns in the state have received Starlink connections to help locals and emergency workers.

The town of Malden was almost completely destroyed, but restoration efforts are underway, and of course it helps to be able to access the internet for communicating with residents and authorities. With power and cellular service unreliable, satellite internet is a good temporary option, and Starlink stepped up.

As SpaceX founder Elon Musk said on Twitter, the company is prioritizing emergency responders and areas without internet:

In a study of gorilla skeletons collected in the wild, Johns Hopkins Medicine researchers and their international collaborators report that aging female gorillas do not experience the accelerated bone loss associated with the bone-weakening condition called osteoporosis, as their human counterparts often do. The findings, they say, could offer clues as to how humans evolved with age-related diseases.

The study was published on Sept. 21, 2020, in Philosophical Translations of the Royal Society B.

“Osteoporosis in humans is a really interesting mechanical problem,” says Christopher Ruff, Ph.D., professor at the Center for Functional Anatomy and Evolution at the Johns Hopkins University School of Medicine. “In terms of natural selection, there is no evolutionary advantage in developing with aging to the point of a potential fracture. By looking at close relatives of humans on the evolutionary tree, we can infer more about the origins of this condition.”

The number of mutations that can contribute to aging may be significantly higher than previously believed, according to new research on fruit flies. The study by scientists at Linköping University, Sweden, supports a new theory about the type of mutation that can lie behind aging. The results have been published in BMC Biology.

We live, we age and we die. Many functions of our bodies deteriorate slowly but surely as we age, and eventually an organism dies. This thought may not be very encouraging, but most of us have probably accepted that this is the fate of all living creatures—death is part of life. However, those who study find it far from clear why this is the case.

“The evolution of aging is, in a manner of speaking, a paradox. Evolution causes continuous adaptation in organisms, but even so it has not resulted in them ceasing to age,” says Urban Friberg, senior lecturer in the Department of Physics, Chemistry and Biology at Linköping University and leader of the study.

autonomous drone delivery

DroneUp and NATO Allied Command Transformation performed an experiment to prove a new and innovative way of resupplying soldiers on the battlefield. The experiment proved that autonomous drone delivery works.

“DroneUp recently partnered with North Atlantic Treaty Organization Allied Command Transformation, Joint Force Development Directorate, Operational Experimentation branch in an experiment designed to determine if autonomous delivery of a specified payload to identified recipients under field conditions could be proven viable,” says a press release.

The experiment took place on September 21, 2020 in Lawrenceville, VA and included Pale Horse

autonomous drone delivery

Weapons Institute, Daniel Defense, Ultimate Training Munitions (UTM), and WeaponLogic. In summary, here’s how the autonomous drone delivery system test worked: soldiers running out of ammunition hit a button (which can be attached to their hat or clothing.) That button signals a drone to fly to that individual soldier and drop a payload – which can be unique to that individual. Then the drone returns home for the next mission.

Digital transformations have become a global trend in recent years. To be clear, in mainstream understanding, the term means to increase the use of data, which can then help us to build “smarter” machines, predict the future, dig out insights, eliminate human errors and maximize efficiency. However, according to the stats released by Boston Consulting Group (BCG) and McKinsey & Company, only about 30% of digital transformation projects ended up successfully. The result keeps us wondering: What are the key issues to account for such high failure rate? And more importantly, how can we resolve these issues?