Lifeboat Foundation

An artificial intelligence system from Google’s sibling company DeepMind stumbled on a new way to solve a foundational math problem at the heart of modern computing, a new study finds. A modification of the company’s game engine AlphaZero (famously used to defeat chess grandmasters and legends in the game of Go) outperformed an algorithm that had not been improved on for more than 50 years, researchers say.

The new research focused on multiplying grids of numbers known as matrices. Matrix multiplication is an operation key to many computational tasks, such as processing images, recognizing speech commands, training neural networks, running simulations to predict the weather, and compressing data for sharing on the Internet.

Motherboard is live, talking with journalist Kelsey Atherton about the use of AI and robotic weapons.

Follow MOTHERBOARD
Facebook: http://www.facebook.com/motherboardtv.
Twitter: http://twitter.com/motherboard.
Tumblr: http://motherboardtv.tumblr.com/
Instagram: http://instagram.com/motherboardvice.
More videos from the VICE network: https://www.fb.com/vicevideo

In this Robotics 24/7 Roundtable, attendees can hear from e-commerce, retail, and omnichannel experts on best practices for selecting and using a variety of robots.

AI image generators, which create fantastical sights at the intersection of dreams and reality, bubble up on every corner of the web. Their entertainment value is demonstrated by an ever-expanding treasure trove of whimsical and random images serving as indirect portals to the brains of human designers. A simple text prompt yields a nearly instantaneous image, satisfying our primitive brains, which are hardwired for instant gratification.

Although seemingly nascent, the field of AI-generated art can be traced back as far as the 1960s with early attempts using symbolic rule-based approaches to make technical images. While the progression of models that untangle and parse words has gained increasing sophistication, the explosion of generative art has sparked debate around copyright, disinformation, and biases, all mired in hype and controversy.

Continue reading “How AI image generators could help robots” »

Over the last three decades, the digital world that we access through smartphones and computers has grown so rich and detailed that much of our physical world has a corresponding life in this digital reality. Today, the physical and digital realities are on a steady course to merging, as robots, Augmented Reality (AR) and wearable digital devices enter our physical world, and physical items get their digital twin computer representations in the digital world.

These digital twins can be uniquely identified and protected from manipulation thanks to crypto technologies like blockchains. The trust that these technologies provide is extremely powerful, helping to fight counterfeiting, increase supply chain transparency, and enable the circular economy. However, a weak point is that there is no versatile and generally applicable identifier of physical items that is as trustworthy as a blockchain. This breaks the connection between the physical and digital twins and therefore limits the potential of technical solutions.

In a new paper published in Light: Science & Applications, an interdisciplinary team of scientists led by Professors Jan Lagerwall (physics) and Holger Voos (robotics) from the University of Luxembourg, Luxembourg, and Prof. Mathew Schwartz (architecture, construction of the built environment) from the New Jersey Institute of Technology, U.S., propose an innovative solution to this problem where physical items are given unique and unclonable fingerprints realized using cholesteric spherical reflectors, or CSRs for short.

For decades researchers have worked to design robotic hands that mimic the dexterity of human hands in the ways they grasp and manipulate objects. However, these earlier robotic hands have not been able to withstand the physical impacts that can occur in unstructured environments. A research team has now developed a compact robotic finger for dexterous hands, while also being capable of withstanding physical impacts in its working environment.

The team of researchers from Harbin University of Technology (China) published their work in the journal Frontiers of Mechanical Engineering on October 14, 2022.

Continue reading “New compact and dexterous robotic finger that can withstand physical impacts” »

Telling artificial intelligence models to “think” step by step when carrying out a task can improve their performance so much that they can outperform humans at jobs AIs usually struggle with.

Researchers have developed a hackable and multi-functional 3D printer for soft materials that is affordable and open design. The technology has the potential to unlock further innovation in diverse fields, such as tissue engineering, soft robotics, food, and eco-friendly material processing—aiding the creation of unprecedented designs.

The findings could help pave the way for greater use of machine learning in materials science, a field that still relies heavily on laboratory experimentation. Also, the technique of using machine learning to make predictions that are then checked in the lab could be adapted for discovery in other fields, such as chemistry and physics, say experts in materials science.

To understand why it’s a significant development, it’s worth looking at the traditional way new compounds are usually created, says Michael Titus, an assistant professor of materials engineering at Purdue University, who was not involved in the research. The process of tinkering in the lab is painstaking and inefficient.