Lifeboat Foundation

Edward Frenkel is a renowned mathematician, professor of University of California, Berkeley, member of the American Academy of Arts and Sciences, and winner of the Hermann Weyl Prize in Mathematical Physics. In this episode, Edward Frenkel discusses the recent monumental proof in the Langlands program, explaining its significance and how it advances understanding in modern mathematics.

SPONSOR (THE ECONOMIST): As a listener of TOE you can get a special 20% off discount to The Economist and all it has to offer! Visit https://www.economist.com/toe.

Continue reading “This is a Monumental Breakthrough… [Part 2]” »

Neural networks have a remarkable ability to learn specific tasks, such as identifying handwritten digits. However, these models often experience “catastrophic forgetting” when taught additional tasks: They can successfully learn the new assignments, but “forget” how to complete the original. For many artificial neural networks, like those that guide self-driving cars, learning additional tasks thus requires being fully reprogrammed.

Biological brains, on the other hand, are remarkably flexible. Humans and animals can easily learn how to play a new game, for instance, without having to re-learn how to walk and talk.

Inspired by the flexibility of human and animal brains, Caltech researchers have now developed a new type of algorithm that enables neural networks to be continuously updated with new data that they are able to learn from without having to start from scratch. The algorithm, called a functionally invariant path (FIP) algorithm, has wide-ranging applications from improving recommendations on online stores to fine-tuning self-driving cars.

It’s clear that 2024 is the year of AI breaking into the world’s most prestigious scientific awards.

Company insists that new machines will be used to help people with their jobs, not take them.

In the last few months, the mask has really come off.

OpenAI is far from its days of being an altruistic, non-profit company.

It is now the face of a booming AI industry and is effectively for-profit in all but name, steaming ahead with little regard for its technology’s environmental toll, or for the potentially existential risks it poses to society.

Continue reading “AI Researcher Slams OpenAI, Warns It Will Become the ‘Most Orwellian Company of All Time’” »

The RUBIK Pi is a dev board from Thundercomm that’s positioned as a platform for developers looking to work a Qualcomm AI processor.

At the heart of the board is a Qualcomm QCS6490 processor with eight ARMv8 CPU cores, Qualcomm Adreno 643 graphics, and a 6th-gen Qualcomm AI Engine that delivers up to 12.5 TOPS of AI performance. Thundercomm hasn’t announced how much the board will cost yet, but says it will be available for pre-order starting in early November.

The Army has sent at least one “robot dog” armed with an artificial intelligence-enabled gun turret to the Middle East for testing as a fresh counter-drone capability for U.S. service members, service officials confirmed.

Photos published to the Defense Visual Information Distribution Service last…

The Army was testing at least one armed quadrupedal unmanned ground vehicle at an installation in Saudi Arabia.

Continue reading “Army Testing Robot Dogs Armed with Artificial Intelligence-Enabled Rifles in Middle East” »

Neural networks have a remarkable ability to learn specific tasks, such as identifying handwritten digits. However, these models often experience “catastrophic forgetting” when taught additional tasks: They can successfully learn the new assignments, but “forget” how to complete the original. For many artificial neural networks, like those that guide self-driving cars, learning additional tasks thus requires being fully reprogrammed.

Biological brains, on the other hand, are remarkably flexible. Humans and animals can easily learn how to play a new game, for instance, without having to re-learn how to walk and talk.

Inspired by the flexibility of human and animal brains, Caltech researchers have now developed a new type of algorithm that enables neural networks to be continuously updated with new data that they are able to learn from without having to start from scratch. The algorithm, called a functionally invariant path (FIP) algorithm, has wide-ranging applications from improving recommendations on online stores to fine-tuning self-driving cars.