philip lelyveld The world of entertainment technology

Scientists Put Flies in ‘Virtual Reality’ to Explore How Vision Shapes Motion

A new study looking at the behavior of fruit flies (Drosophila melanogaster) in a 'virtual reality' setting offers up some clues – and it seems that conventional scientific wisdom on how vision and movement intertwine might be wrong.

The new experiments show that vision was used to prevent the flies from going off their intended course before it happened, rather than to get them back on track after they'd already deviated from a route – and that's a substantial difference.

...By treating the postural signals coming from elsewhere as less important when they could see where they were going, flies appeared to use their vision to preemptively keep their bodies on course. That suggests a very close link between sight and motor control.

See the full story here: https://www.sciencealert.com/scientists-put-fruit-flies-in-virtual-reality-to-see-how-their-vision-affects-locomotion

Rendever unveils its ‘RendeverFit’ Virtual Reality fitness platform for senior living and long-term care

Rendever, a provider of a virtual reality (VR) platform built to help seniors overcome social isolation through shared experiences, has announced the launch of its new interdisciplinary VR fitness platform, ‘RendeverFit’ – a VR application that merges physical fitness with cognitive fitness and socialization, designed specifically for older adults.

RendeverFit was created to combine physical activity with social engagement for senior living residents with physical activity and social engagement for older adults, both of which can help to improve overall physical and mental wellbeing, according to Rendever.

See the full story here: https://www.auganix.org/rendever-unveils-its-rendeverfit-virtual-reality-fitness-platform-for-senior-living-and-long-term-care/

National committee will advise the President on AI competition and ethics

The Biden administration's focus on science will include a strong emphasis on artificial intelligence. The Commerce Department, National AI Initiative Office and White House Office of Science and Technology Policy are forming a National Artificial Intelligence Advisory Committee (NAIAC) to advise the President and federal officials on AI-related issues.

NAIAC will provide guidance on several AI concerns, including "competitiveness," employment, scientific progress, the viability of national strategy and future initiative revisions. The committee will also address ethical issues ranging from workforce equity to accountability and algorithmic bias.

Members will come from a "broad and interdisciplinary" pool including academics, companies, non-profits and federal labs. They'll be accepted on a continuous basis, with any eventual vacancies filled as they pop up.

See the full story here: https://www.yahoo.com/entertainment/commerce-department-national-ai-advisory-committee-172406939.html

Australian media outlets liable for Facebook comments, court finds

Australia's biggest news publishers including Rupert Murdoch's the Australian are responsible for comments that readers post on their corporate Facebook pages, the High Court ruled on Wednesday.

"The acts of the (media companies) in facilitating, encouraging and thereby assisting the posting of comments by the third-party Facebook users rendered them publishers of those comments," Justice Rothman found.

See the full story here: https://www.reuters.com/business/media-telecom/australian-media-outlets-liable-facebook-comments-court-finds-2021-09-08/

Federal Judge Rules AI-Enabled Machines Are Not Inventors

U.S. District judge Leonie Brinkema just ruled that an artificial intelligence-enabled computer cannot be listed as an inventor on patents and that only humans can be inventors under U.S. law. That’s because, according to Federal law, an “individual” must take an oath that he/she is the inventor and the term “individual” is legally defined as a natural person. The ruling was in response to University of Surrey law professor Ryan Abbott’s effort, the Artificial Inventor Project, to get a computer listed as an inventor.

https://www.etcentric.org/federal-judge-rules-ai-enabled-machines-are-not-inventors/

and

https://www.bloomberg.com/news/articles/2021-09-03/only-humans-not-ai-machines-can-get-a-u-s-patent-judge-rules

Are Blockchain Networks Living Organisms?

IN BRIEF

• Blockchain networks meet most of the traditional criteria for life.
• Their combination with artificial neural network (ANN) based artificial intelligence (AI) could lead to the creation of a distributed blockchain organism.
• Such a cybernetic organism could exhibit unique features: directed evolution, the ability to self-improve or an unlimited lifespan.

Conclusion

On the surface, calling a virtual blockchain network based on pure numerical relations a living organism seems absurd. However, on closer examination, it turns out that it meets most of the traditional criteria of what is alive and organic.

The obvious fact that blockchain is not a biological organism need not be seen as an imperfection. On the contrary, its non-material components represent the potential advantage and progress that blockchain technology can bring to the understanding and applications of such traditional concepts as an information carrier, evolution, inheritance, emergence, or self-organization.

Harnessing the unique properties of blockchain networks and combining them with artificial intelligence based on artificial neural networks could lead to a breakthrough. 

The result could be a unique organism – a public, blockchain-based, distributed virtual machine. It would potentially be capable of directed evolution, rapid self-improvement, and unlimited lifespan. Alternatively, it would go far beyond human cognitive abilities and intelligence.

See the full story here: https://beincrypto.com/are-blockchain-networks-living-organisms/

USE VIRTUAL REALITY TO PREPARE MARITIME CREWS FOR TERRORIST AND PIRACY ATTACKS

In the maritime industry, the unexpected and ongoing attacks at sea require a different way of thinking and a different point of view on safety and procedures. Already, maritime simulators have become one of the most advance forms of education across the world. Even the most experienced professionals are regularly trained in such simulators on emergency operations, demonstrating a renewed emphasis on operator training. VR-training software is an efficient and easy alternative to face-to-face training.

Selina Robinson is a Teaching Fellow in Forensic Investigations at the University of Portsmouth. Her areas of research lie with maritime security in cruise ships and the use of XR simulations in education and training. Previous work experience involves counter terrorism and crime scene investigation. 

Dr. Amy Meenaghan is a lecturer in Psychology in the School of Criminology and Criminal Justice Studies at the University of Portsmouth. Her background is in the use of virtual reality to understand offender decision making. She is currently working on various VR projects within the field of Criminal Justice, with a focus on optimizing such technologies to benefit education, training and crime prevention initiatives.

See the full story here: https://cimsec.org/use-virtual-reality-to-prepare-maritime-crews-for-terrorist-and-piracy-attacks/

Avegant Introduces The World’s Smallest LED Light Engines For Augmented Reality

Avegant today announced its latest products -- the AG-30L 30º and AG-50L 50º LED light engines. Avegant's 30º AR light engine uses proprietary illumination technology to dramatically reduce engine volume and weight while preserving efficiency and MTF. The next generation Avegant light engine, the AG-50L, enables a larger field of view and a more immersive AR experience for users while maintaining the compact, light glasses form factor made possible by the AG-30L product.

A limited supply of engineering samples of Avegant's AG-30L light engine will be available in late 2021. Engineering samples of Avegant's AG-50L light engine will be available in early 2022. Mass production is anticipated to begin in 2023.

See the full story here: https://www.prnewswire.com/news-releases/avegant-introduces-the-worlds-smallest-led-light-engines-for-augmented-reality-301370226.html

VIRTUAL REALITY PUTS DRUG RESEARCHERS INSIDE THE MOLECULES THEY STUDY

Drug discovery is less like finding a key to fit a lock—a familiar but misleading analogy—and more like finding a dance partner who can match your moves, as scientists at U.K.-based biotech company C4X Discovery Holdings PLC put it.

C4X is designing small molecules to find drugs for treating cancer and autoimmune diseases. The company is developing its own VR platform, 4Sight, to help make drug-molecule visualization faster and easier.

Nanome, which was founded by experts in nanoengineering, electrical engineering, human-computer interaction and VR, is working to develop an AI assistant that can guide users through their VR experience. Its virtual assistant could ask scientists, for example, if they want to see more or less atomic detail or additional chemical interactions. 

For now, Dr. von Salm says she is pleased to stand in a virtual room and visualize her compounds without having to rely so much on imagination or computer renderings.

“Even for the best people who have a really amazing imagination, you’d put them in this room, and they’d be in awe,” she says.

See the full story here: https://www.wsj.com/articles/virtual-reality-puts-drug-researchers-inside-the-molecules-they-study-11631023212

4 Things GPT-4 Will Improve From GPT-3

... GPT-3 was brute-force trained in most of the Internet’s available text data. But it wasn’t explicitly taught to do these tasks. The system is so powerful that it became a meta-learner. It learned how to learn. And users could communicate with it in plain natural language; GPT-3 would receive the description and recognize the task it had to do. ...

Disclaimer: GPT-4 doesn’t exist (yet). ...

GPT-3 is big. GPT-4 will be bigger. Here’s the reason... GPT-3 isn’t just more powerful than GPT-2, it is differently more powerful. There’s a qualitative leap between both models. ...

GPT-4 will perform better few-shot multitasking... GPT-4 could be the proof that language models can learn multitasking from a few examples almost as good as we can. ...

GPT-4 won’t depend as much on good prompting ... Tech blogger Gwern Branwen ... calls this way of interacting with GPT-3 prompt programming. We can give GPT-3 a written input and it’ll know which task it has to perform. ... However, Gwern warns that the results of these tests can vary in quality. The reason is that prompt programming involves sampling. And sampling can “prove the presence of knowledge but not the absence”....

GPT-4 will have a larger context window...The beta API allows the user to input a text 500–1000 words long — the context window — for GPT-3 to do something with it. This means the system can’t continue a half-written novel or complete the code for a large program. ...

See the full story here: https://towardsdatascience.com/4-things-gpt-4-will-improve-from-gpt-3-2b1e7a6da49f