Posted in futurism
The future of Foundation Models will be embodied agents that proactively take actions, endlessly explore the world, and continuously self-improve. What does it take? In our NeurIPS Outstanding Paper “MineDojo”, we provide a blueprint for this future:🧵”.
Posted in biotech/medical, cyborgs, genetics, science
An advanced optogenetic visual prosthesis for patients with serious blindness due to photoreceptor loss.
Human behaviour is remarkably complex. Even a simple request like, “Put the ball close to the box” still requires deep understanding of situated intent and language. The meaning of a word like ‘close’ can be difficult to pin down – placing the ball inside the box might technically be the closest, but it’s likely the speaker wants the ball placed next to the box. For a person to correctly act on the request, they must be able to understand and judge the situation and surrounding context.
Most artificial intelligence (AI) researchers now believe that writing computer code which can capture the nuances of situated interactions is impossible. Alternatively, modern machine learning (ML) researchers have focused on learning about these types of interactions from data. To explore these learning-based approaches and quickly build agents that can make sense of human instructions and safely perform actions in open-ended conditions, we created a research framework within a video game environment.
Today, we’re publishing a paper and collection of videos, showing our early steps in building video game AIs that can understand fuzzy human concepts – and therefore, can begin to interact with people on their own terms.
Neuralink’s invasive brain implant vs phantom neuro’s minimally invasive muscle implant. Deep dive on brain computer interfaces, Phantom Neuro, and the future of repairing missing functions.
Connor glass.
Phantom is creating a human-machine interfacing system for lifelike control of technology. We are currently hiring skilled and forward-thinking electrical, mechanical, UI, AR/VR, and Ai/ML engineers. Looking to get in touch with us? Send us an email at [email protected].
Phantom Neuro.
Phantom is a neurotechnology company, spun out of the lab at The Johns Hopkins University School of Medicine, that is enabling lifelike control of robotic orthopedic technologies, such as prosthetic limbs and exoskeletons. Phantom’s solution, the Phantom X, consists of low-risk implantable sensors, AI, and enabling software. By providing superior control of robotic orthopedic mechanisms, the Phantom X will drastically improve the lives of individuals with limb difference who have yet to see a tangible improvement in quality of life despite significant advancements in the field of robotics.
Links:
Posted in futurism
How many types of leukemia are there? What makes each one unique, and are they treated any differently?
We went to leukemia specialist Naveen Pemmaraju, M.D., for answers to these questions and more. Here’s what he shared.
Unnecessary playing with nature.
In January, Bennett’s doctors offered him the chance to receive a heart from a pig. He took it. “I know it’s a shot in the dark, but it’s my last choice,” he said in a press release from the University of Maryland Medical Center in Baltimore, where he was being treated. On 7 January, doctors transplanted the heart, which had been genetically modified so that the human body would tolerate it.
Bennett survived for eight weeks with his new heart before his body shut down. After his death, the research team learnt that the transplanted organ was infected with a pig herpesvirus that had not been detected by tests1.
But even a few weeks is a long time for an animal organ placed in a human, known as a xenotransplant. Given that the human immune system begins attacking non-genetically modified pig organs in minutes, other xenotransplantation researchers are impressed with the experiment. “It’s actually beyond my expectation that the patient lived up to two months,” says Luhan Yang, a bioengineer and chief executive of Qihan Biotech in Hangzhou, China. “I think it’s a victory for the field.”
Imagine that you can grow a clone with the level of intelligence below the level of a turkey or a chicken with the perfectly-exercised body. Would you consider swapping your brain with this perfect and unconscious clone? Surprisingly, many professionals would not even think twice.
Scientists in Israel have created the first nano-robot antibodies designed to fight cancer. The first human trial for the new nano-robots will start soon, and it will determine just how effective the antibodies are. What is special about these particular antibodies, too, is that they are programmed to decide whether cells surrounding tumors are “bad” or “good.”
The trial is currently underway in Australia and if it goes according to plan, the nano-robot antibodies will be able to fight cells around tumors that can help the tumor while also boosting the capability of the cells inhibiting the growth of the cancerous cells. The antibodies were invented by Professor Yanay Ofran and are based on human and animal antibodies.
The goal of these nano-robot antibodies is to unlock the full potential that antibodies offer, Ofran says. Currently, the use of antibodies in medicine only utilizes a fraction of the capabilities offered by these natural disease fighters. As such, finding a way to maximize their capability has been a long-term goal for quite a while.
A new technique enables artificial intelligence agents to think much farther into the future when considering how their behaviors can influence the behaviors of other AI agents, toward the completion of a task. This approach improves long-term performance of cooperative or competitive AI agents.
A recent study, affiliated with South Korea’s Ulsan National Institute of Science and Technology (UNIST) has reported a scalable synthetic strategy to fabricate low-resistance edge contacts to atomic transistors using a thermally stable 2D metal, namely PtTe2.
Developing cheaper, smaller, and better-performing semiconductors with materials other than silicon (Si), is expected to gain momentum, thanks to a recent study from UNIST. This will aid in reducing the space between semiconductors and metals within semiconductor devices to ∼1 nm, which could help maintain high performance.
Published in the August 2022 issue of Nature Communications, this study has been jointly led by Professor Soon-Yong Kwon and Professor Zonghoon Lee in the Department of Materials Science and Engineering at UNIST.
