Sushi Nooz Special: SINGULARITY MINDS Part 2

Author: Sushi Nooz

June 28, 2021 

Here's the link to Part 1.

1. Artificial General Intelligence is here, DeepMind unveils its Impala AGI – By Futurist and Virtual Keynote Speaker Matthew Griffin
   
   
2. Pricing algorithms can learn to collude with each other to raise prices - MIT Technology Review
   
   
3. MIT's new neural network can explain its decisions – Fanatical Futurist by International Keynote Speaker Matthew Griffin
   
   
4. AI develops human-like number sense – taking us a step closer to building machines with general intelligence
   
   
5. MIT CSAIL's AI can visualize objects using touch | VentureBeat
   
   
6. DeepMind Teaches AI Teamwork - IEEE Spectrum
   
   
7. Robot arm learns how to taste with engineered bacteria
   
   
8. MIT's AI model learns relationships among objects with minimal training data
   
   
9. Google uses AI to teach a robot how to grasp and throw things | VentureBeat
   
   
10. Researchers propose framework that trains robots to mimic human actions in captioned videos | VentureBeat
    
    
11. Watch a Self-Driving Car Careen Around Corners Like a Racecar
    
    
12. AI can predict when someone will die with unsettling accuracy
    
    
13. A Prominent Publisher Used Machine Learning to Write a Textbook
    
    
14. OpenAI Five defeats professional Dota 2 team, twice | VentureBeat
    
    
15. Artificial intelligence singles out neurons faster than a human can
    
    
16. AI Robot paints its own moonscapes in traditional Chinese style - Reuters
    
    
17. This clever AI hid data from its creators to cheat at its appointed task – TechCrunch
    
    
18. Emotionally intelligent AI will respond to how you feel
    
    
19. Watch "Full Self-Driving" on YouTube
    
    
20. Magnets aid AI achieve efficiency of the human brain
    
    
21. IBM’s New AI Does Something Amazing: It Learns From “Memories”
    
    
22. MIT's AI makes autonomous cars drive more like humans | VentureBeat
    
    
23. Astounding AI Guesses What You Look Like Based on Your Voice
    
    
24. Facebook’s AI learns how to get around an office by watching videos
    
    
25. Tech & Science Archives - Digital Journal
    
    
26. This AI Watched 100 Films to Learn How to Recognize a Kiss - IEEE Spectrum
    
    
27. AI classifies people's emotions from the way they walk | VentureBeat
    
    
28. AI's Minority Report for retail: They know you’ll return it even before you buy it | ZDNet
    
    
29. Facebook AI Pluribus defeats top poker professionals in 6-player Texas Hold ’em
    
    
30. AI Pores Over Old Scientific Papers, Makes Discoveries Overlooked By Humans | Zero Hedge
    
    
31. First human drug developed solely by AI is a vaccine
    
    
32. Deep learning algorithm solves Rubik's Cube faster than any human
    
    
33. AI passes theory of mind test by imagining itself in another's shoes | New Scientist
    
    
34. Alphabet's Loon AI-powered internet balloons have learned to tack like a sailor and move with the wind | ZDNet
    
    
35. Tencent's Honor of Kings AI beats a team of pros | VentureBeat
    
    
36. AI learns to predict the outcomes of human rights court cases | New Scientist
    
    
37. A startup that marries AI with empathy is helping women conceive | MIT Technology Review
    
    
38. The brain inspires a new type of artificial intelligence -- ScienceDaily
    
    
39. Amazon says its facial recognition can now identify fear
    
    
40. MIT Model Automates AI for Medical Decision-Making

Sushi Nooz Special: SINGULARITY MINDS

Author: Sushi Nooz

June 28, 2021 

 

Here's the link to Part 2

Here are some of the links that I have been putting together over the years on AGI (Artificial General Intelligence).

I have been doing research for many years and would like to show you, the reader, what I have found on a variety of topics. This topic of AGI, AI, is something that I find very fascinating and I have carefully curated the following links for your viewing pleasure and to help more people be aware of what's going on, and find information that is not being shared on traditional websites and news publications.

1. How a New AI Model Mimics the Brain’s Prefrontal Cortex | Psychology Today
   
   
2. Stanford researchers propose AI that figures out how to use real-world objects | VentureBeat
   
   
3. AI models from Microsoft and Google already surpass human performance on the SuperGLUE language benchmark | VentureBeat
   
   
4. Stanford researchers propose AI that figures out how to use real-world objects | VentureBeat
   
   
5. The AI-Powered Robot That Learnt Curling Using Adaptive Deep Reinforcement Learning
   
   
6. AI devs claim they've built a robot that shows a 'primitive form of empathy'
   
   
7. Tweaking AI software to function like a human brain improves computer's learning ability -- ScienceDaily
   
   
8. Can AI Machine Learning Enable Robot Empathy? | Psychology Today
   
   
9. AI Gains Social Intelligence; Infers Goals and Failed Plans | Psychology Today
   
   
10. New AI Outperforms State-of-the-Art Machine Hearing | Psychology Today
    
    
11. This AI created its own lesson plans to learn faster then smashed other AI's out the park – By Futurist and Virtual Keynote Speaker Matthew Griffin
    
    
12. AI smashes video game high scores by remembering its past success | New Scientist
    
    
13. Reinforcement learning algorithms score higher than humans, other AI systems at classic video games
    
    
14. Neural network CLIP mirrors human brain neurons in image recognition
    
    
15. 'Self-trained' deep learning to improve disease diagnosis
    
    
16. Algorithm helps artificial intelligence systems dodge 'adversarial' inputs
    
    
17. This robot artist stops to consider its brushstrokes like a real person | Engadget
    
    
18. AI Can Now Debate with Humans and Sometimes Convince Them, Too - Scientific American
    
    
19. To Learn To Deal With Uncertainty, This AI Plays Pong - IEEE Spectrum
    
    
20. After AIs mastered Go and Super Mario, scientists have taught them how to
    'play' experiments at NSLS-II
    
    
21. Scientists Create 'Living' Robots That Think And Have Memories With Frog Cells
    
    
22. Artificial nervous system senses light and learns to catch like humans | New Scientist
    
    
23. ‘You Look Good Today. Want Snacks?’: Artificial Intelligence Learns To Flirt Thanks To Colorado Scientist Janelle Shane – CBS Denver
    
    
24. To advance AI, Facebook is leaning into medical research
    
    
25. ‘Brain-like device’ mimics human learning in major computing breakthrough
    
    
26. AI spots individual neurons in the eye better than human experts
    
    
27. Self-learning robots go full steam ahead
    
    
28. Artificial intelligence programme that detects sarcasm in social media | Technology News | Zee News
    
    
29. AI that mimics human typos on a smartphone could improve keyboards | New Scientist
    
    
30. Facebook has created a way for AI, like people, to forget unimportant details it learns - SiliconANGLE
    
    
31. Increasing the memory capacity of intelligent systems based on the function of human neurons

Researchers create a cyborg rat that can be controlled with human mind control

(Source: fanaticalfuturist.com)

8th June 2021

Matthew Griffin

WHY THIS MATTERS IN BRIEF

We are slowly unlocking the mysteries of the mind – and now we’re starting to explore the limits …

Everyone loves rats – even if they’re “Franken-Rats” with two heads like the rats from this head transplant experiment. Or share a hive mind that lets little cuddly critters on different continents learn from one another via the internet. And now scientists have created human-controlled rat cyborgs. Lest you think this is some media sensationalism at work, here’s the actual title of the paper under discussion, which came out last week in Scientific Reports: “Human Mind Control of Rat Cyborg’s Continuous Locomotion with Wireless Brain-to-Brain Interface.”

That pretty much says it all. Some of this tech – such as Brain to Brain Interfaces (BBIs) – is nothing new in science, so in a way this just a small step in an already existing race. But, put another way: people are controlling rat cyborgs with their freaking brains now. Wirelessly.

Here’s the deal. BBIs are themselves based on stringing together BMIs, or Brain Machine Interfaces. These already exist, and present a cool way for people to control prosthetics or other devices. Apparently they can also function the other way around — instead of a brain controlling a device, a machine can alter brain patterns or “import tactile information back to the brain,” as the study’s authors put it.

So, BMIs allow for mechanically “controlling” others’ brains. In effect, it works like this: A human has movement-related thoughts, which an EEG picks up and transfers to a computer. The computer translates that signal into “control instructions,” which get wirelessly beamed into the stimulator on the back of the rat and then into its brain via electrodes – which is, by the way, now a rat cyborg because of its cybernetic parts.

The rat then responds to the instructions by actually doing them. All this tech exists, but for some reason “very few previous studies have explored BBIs across different brains,” the authors write — so they fixed that with a very real, actual experiment that involved humans wirelessly controlling rat cyborgs through mazes. I can’t stress enough how real this is, or how often the authors used the phrase “rat cyborg” in their peer-reviewed scientific paper.

The researchers, all from Zhejiang University in China, explain that they wanted a system that worked nearly instantaneously and with high accuracy, due to a living creatures’ “agility and self-consciousness.” And, apparently, that’s what they got.

“With this interface, our manipulators were able to mind control a rat cyborg to smoothly complete maze navigation tasks,” the authors write, almost gloating. The human controller was non-invasively hooked into the BBI, where thinking about moving their left or right arms would command the rat to turn in those directions, and blinking corresponded to forward.

The study used six rat cyborgs, well-trained and previously prepared for such control. The final part of the system involved a camera showing the rat’s real-time reactions, allowing the human controller to get instant visual feedback. At first the maze was simple, a multi-armed set of tubes that met in the middle, like the spokes of a wheel. The idea was to use it to practice commanding the rat through turns. Then the mazes got more complex, with tight turns, multiple levels and a specific prescribed path. The rat cyborgs handled well, overall, with improved control over time; two of the rats apparently performed flawlessly. For their parts, the human controllers were subject to fatigue and easily distracted, but over time they improved, too, cultivating what the writers call “a tacit understanding between the human and the rat cyborg.”

I know what you’re thinking – a chilling phrase in light of the subject matter: Is this for real?

“The Zhang paper is for real, and I don’t see anything implausible about it,” says University of Washington brain researcher Andrea Stocco, who wasn’t involved with the work.

“The results … are impressive but believable.” As for next steps, the study’s authors discuss ways to improve the technology, but notably don’t suggest any particular applications, perhaps leaving those as an exercise to the reader. They do call the whole idea of BBI transmitting information between two entities “intriguingly possible,” and add, “furthermore, information flow will be made bidirectional and communicative between two human individuals.” Which then gets us closer to being able to create human hive minds… but that’s another story.

Stocco suggests the tech could be involved in improved augmented reality systems, remote control of an animal through tight spaces — or even an expert surgeon remotely controlling a doctor’s hands in a delicate operation.

Ultimately, he says, “the holy grail of BBI would be sharing rich content that cannot be better expressed in words, such as emotions and feelings. We are still so far from that, but, of course, that would be the dream.”

For his part, Harvard lab director Seung-Schik Yoo, who is also unaffiliated with the paper, imagines such technology could benefit people even more directly.

“I personally want to use brain stimulation [for its] neurotherapeutic applications,” he says, adding “I am one of the believers who think that brain stimulators will be part of our daily lives as ‘neuroprostheses’ some day. Of course, we have to fear the misuse of the technology and prevent it from happening. Many new technologies are a double-edged sword.” But Stocco isn’t too worried.

“We are very, very far from any application of these technologies on humans against their will,” he says. And anyway, it’s not like we need BBI to manipulate people even now. “Scandals like Cambridge Analytica remind us that social media feeds and advertisement campaigns are just the good old-fashioned way to achieve the same goals of making other people do something they wouldn’t do otherwise.”

So really, this research is just bringing up some of the same old ethical dilemmas we should already be used to dealing with. Only, you know, featuring rat cyborgs.