Toggle light / dark theme

DeepMind AI achieves gold-medal level performance on challenging Olympiad math questions

A team of researchers at Google’s DeepMind project, reports that its AlphaGeometry2 AI performed at a gold-medal level when tasked with solving problems that were given to high school students participating in the International Mathematical Olympiad (IMO) over the past 25 years. In their paper posted on the arXiv preprint server, the team gives an overview of AlphaGeometry2 and its scores when solving IMO problems.

Prior research has suggested that AI that can solve geometry problems could lead to more sophisticated apps because they require both a high level of reasoning ability and an ability to choose from possible steps in working toward a solution to a problem.

To that end, the team at DeepMind has been working on developing increasingly sophisticated geometry-solving apps. Its first iteration was released last January and was called AlphaGeometry; its second iteration is called AlphaGeometry2.

DeepMind claims its AI performs better than International Mathematical Olympiad gold medalists

An AI system developed by Google DeepMind, Google’s leading AI research lab, appears to have surpassed the average gold medalist in solving geometry problems in an international mathematics competition.

The system, called AlphaGeometry2, is an improved version of a system, AlphaGeometry, that DeepMind released last January. In a newly published study, the DeepMind researchers behind AlphaGeometry2 claim their AI can solve 84% of all geometry problems over the last 25 years in the International Mathematical Olympiad (IMO), a math contest for high school students.

Why does DeepMind care about a high-school-level math competition? Well, the lab thinks the key to more capable AI might lie in discovering new ways to solve challenging geometry problems — specifically Euclidean geometry problems.

Generalizing Safety Beyond Collision-Avoidance via Latent-Space Reachability Analysis

Hamilton-Jacobi (HJ) reachability is a rigorous mathematical framework that enables robots to simultaneously detect unsafe states and generate actions that prevent future failures. While in theory, HJ reachability can synthesize safe controllers for nonlinear systems and nonconvex constraints.

In practice, it has been limited to hand-engineered collision

Avoidance constraints modeled via low-dimensional state-space representations and first-principles dynamics. In this work, our goal is to generalize safe robot controllers to prevent failures that are hard—if not impossible—to write down by hand, but can be intuitively identified from high-dimensional observations:

Google Launches Gemini 2.0 Pro LLM

In today’s AI news, Google launched its much-anticipated new flagship AI model, Gemini 2.0 Pro Experimental, on Wednesday. The announcement was part of a series of other AI model releases. The company is also making its reasoning model, Gemini 2.0 Flash Thinking, available in the Gemini app.

In other advancements, LinkedIn is testing a new job-hunting tool that uses a custom large language model to comb through huge quantities of data to help people find prospective roles. The company believes that artificial intelligence will help users unearth new roles they might have missed in the typical search process.

S Deep Research feature, which can autonomously browse the web and create research reports. ‘ + s up from hitting $50 million ARR, or the yearly value of last month s case for why they are the best positioned to take over TikTok And, in this episode, a16z Partner Marc Andrusko chats with Mastercard’s Chief AI and Data Officer Greg Ulrich about Mastercard’s long history of using AI, the opportunities (and potential risks) associated with integrating generative AI into fraud detection, determining what tech to employ based on use cases, and the best advice he’s ever gotten.

Then, power your AI transformation with an insightful keynote from Scott Guthrie, Executive Vice President, Cloud + AI Group at Microsoft, and other industry experts. Watch this keynote presentation from NYC stop on Microsoft’s AI Tour.

We close out with this insightful discussion with Malcolm Gladwell and Ric Lewis, SVP of Infrastructure at IBM. Learn how hardware capabilities enable the matrix math behind large language models and how AI is transforming industries—from banking to your local coffee shop.

Thats all for today, but AI is moving fast — like, comment, and subscribe for more AI news! Please vote for me in the Entrepreneur of Impact Competition today! Thank you for supporting my partners and I — it’s how I keep Neural News Network free.

[](https://open.substack.com/pub/remunerationlabs/p/google-laun…hare=true)

New AI Tool Detects Fake News with 99% Accuracy

A tool developed by Keele University researchers has been shown to help detect fake news with an impressive 99% level of accuracy, offering a vital resource in combating online misinformation.

The researchers Dr. Uchenna Ani, Dr. Sangeeta Sangeeta, and Dr. Patricia Asowo-Ayobode from Keele’s School of Computer Science and Mathematics, used a number of different machine learning techniques to develop their model, which can scan news content to give a judgment of whether a news source is trustworthy and genuine or not.

The method developed by the researchers uses an “ensemble voting” technique, which combines the predictions of multiple different machine learning models to give an overall score.

Barbecue grill approach helps researchers understand puzzling Rayleigh–Bloch waves

So-called Rayleigh–Bloch waves can release an enormous amount of energy that can damage technical systems under certain circumstances. They only exist below a precisely defined cut-off frequency; above this, they disappear abruptly. Strangely enough, however, there are isolated high frequencies at which they can also be detected.

Mathematicians from the Universities of Augsburg and Adelaide have recently proposed an explanation for this puzzling phenomenon. Together with researchers from the University of Exeter, they have now been able to prove experimentally that their theory is indeed correct. The study has just been published in the journal Communications Physics.

Suppose you had a gigantic barbecue grill that could easily accommodate several hundreds of sausages. Then, you could not only use it to invite your children’s entire school to a barbecue. The numerous stainless steel struts aligned parallel to each other are also ideal for generating Rayleigh–Bloch waves.

Quantum Systems Obey Second Law Of Thermodynamics

The fundamental principles of thermodynamics have long been a cornerstone of our understanding of the physical world, with the second law of thermodynamics standing as a testament to the inexorable march towards disorder and entropy that governs all closed systems. However, the realm of quantum physics has traditionally appeared to defy this notion, with mathematical formulations suggesting that entropy remains constant in these systems.

Recent research has shed new light on this seeming paradox, revealing that the apparent contradiction between quantum mechanics and thermodynamics can be reconciled through a nuanced understanding of entropy itself. By adopting a definition of entropy that is compatible with the principles of quantum physics, specifically the concept of Shannon entropy, scientists have demonstrated that even isolated quantum systems will indeed evolve towards greater disorder over time, their entropy increasing as the uncertainty of measurement outcomes grows.

This breakthrough insight has far-reaching implications for our comprehension of the interplay between quantum theory and thermodynamics, and is poised to play a pivotal role in the development of novel quantum technologies that rely on the manipulation of complex many-particle systems.