Lifeboat Foundation

Washington (AFP) — Facebook unveiled an initiative Tuesday to take on “hateful memes” by using artificial intelligence, backed by crowd sourcing, to identify maliciously motivated posts.

The leading social network said it had already created a database of 10,000 memes — images often blended with text to deliver a specific message — as part of a ramped-up effort against hate speech.

“These efforts will spur the broader AI research community to test new methods, compare their work, and benchmark their results in order to accelerate work on detecting multimodal hate speech,” Facebook said in a blog post.

Stroke is the leading cause of serious long-term disability in the US with approximately 17 million individuals experiencing it each year. About 8 out of 10 stroke survivors suffer from “hemiparesis”, a paralysis that typically impacts the limbs and facial muscles on one side of their bodies, and often causes severe difficulties walking, a loss of balance with an increased risk of falling, as well as muscle fatigue that quickly sets in during exertions. Oftentimes, these impairments also make it impossible for them to perform basic everyday activities.

To allow stroke patients to recover, many rehabilitation centers have looked to robotic exoskeletons. But although there are now a range of exciting devices that are enabling people to walk again who initially were utterly unable to do so, there remains significant active research trying to understand how to best apply wearable robotics for rehabilitation after stroke. Despite the promise, recent clinical practice guidelines now even recommend against the use of robotic therapies when the goal is to improve walking speed or distance.

In 2017, a multidisciplinary team of mechanical and electrical engineers, apparel designers, and neurorehabilitation experts at Harvard’s Wyss Institute for Biologically Inspired Engineering and John A. Paulson School of Engineering and Applied Sciences (SEAS), and Boston University’s (BU) College of Health & Rehabilitation Sciences: Sargent College showed that an ankle-assisting soft robotic exosuit, tethered to an external battery and motor, was able to significantly improve biomechanical gait functions in stroke patients when worn while walking on a treadmill. The cross-institutional and cross-disciplinary team effort was led by Wyss faculty members Conor Walsh, Ph.D. and Lou Awad, P.T., D.P.T., Ph.D, together with Terry Ellis, Ph.D., P.T., N.C.S. from BU.

In the movie “Transformers,” cars morph into robots, jets or a variety of machinery. A similar concept inspired a group of researchers to combine gas foaming, which is a blend of chemicals that induces gas bubbling, and 3D molding technologies to quickly transform electrospun membranes into complex 3D shapes for biomedical applications.

In Applied Physics Reviews, the group reports on its new approach that demonstrates significant improvements in speed and quality compared with other methods. The work is also the first successful demonstration of formation of 3D neural tissue constructs with an ordered structure through differentiation of human neural progenitor/stem cells on these transformed 3D nanofiber scaffolds.

“Electrospinning is a technology to produce nanofiber membranes,” said co-author Jingwei Xie, at the University of Nebraska Medical Center. “The physics principle behind it involves applying an electrical force to overcome the surface tension of a solution to elongate a solution jet into continuous and ultrafine fibers after solvent evaporation.”

No one has ever seen any airplane like this, except on computer animation. Now, some of the world’s top aeronautical engineers are going to build it for real.

The plan calls for constructing a six-ton unmanned, remote controlled plane the size of a business jet with 24 spinning propellers embedded in its huge moveable wings that allow it to magically hover in midair.

It’s an experimental airplane they call LightningStrike.

A growing cadre of startups is pursuing iterative cycles of machine learning, wet-lab experimentation and human feedback to accelerate target drug discovery.

A new AI diagnostic tool trained on crowdsourced symptom data can predict whether someone likely has COVID-19 without testing.

The model was trained on data from more than 2.5 million users of the COVID Symptom Study app developed at King’s College London, which anyone can download to report their daily health status.

Conclusion

The race is on to develop the hardware that will power the upcoming era of AI. More innovation is happening in the semiconductor industry today than at any time since Silicon Valley’s earliest days. Untold billions of dollars are in play.

This next generation of chips will shape the contours and trajectory of the field of artificial intelligence in the years ahead. In the words of Yann LeCun: “Hardware capabilities…motivate and limit the types of ideas that AI researchers will imagine and will allow themselves to pursue. The tools at our disposal fashion our thoughts more than we care to admit.”

Motion picture animation and video games are impressively lifelike nowadays, capturing a wisp of hair falling across a heroine’s eyes or a canvas sail snapping crisply in the wind. Collaborators from the University of California, Los Angeles (UCLA) and Carnegie Mellon University have adapted this sophisticated computer graphics technology to simulate the movements of soft, limbed robots for the first time.

Next-generation VTOL concepts are rising to meet the future needs of a modern-day battlefield.

Vertical take-off and landing (VTOL) concepts for unmanned aerial systems (UAS) certainly aren’t new. Their reconnaissance and intelligence-gathering roles date back to the 1950s, and there’s been a gradual path toward technological advancements in the decades since.