{"id":296319,"date":"2026-02-17T22:06:09","date_gmt":"2026-02-17T22:06:09","guid":{"rendered":"https:\/\/www.newsbeep.com\/il\/296319\/"},"modified":"2026-02-17T22:06:09","modified_gmt":"2026-02-17T22:06:09","slug":"computer-chips-designed-like-biological-brains-can-finally-handle-massive-math-problems-without-guzzling-energy-like-a-normal-supercomputer","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/il\/296319\/","title":{"rendered":"Computer Chips Designed Like Biological Brains Can Finally Handle Massive Math Problems Without Guzzling Energy Like a Normal Supercomputer"},"content":{"rendered":"

\"\"<\/a>AI-generated illustration. Credit: ZME Science\/Nanobanana.<\/p>\n

When you swing a tennis racket or catch a set of keys, you aren\u2019t thinking about wind resistance or gravity. Yet, to perform that motion, your brain is solving a massive physics problem in milliseconds. It is processing the same kind of complex math that typically demands a warehouse-sized supercomputer.<\/p>\n

Researchers Brad Theilman and James Aimone from Sandia National Laboratories have now demonstrated that neuromorphic hardware can bridge the gap between the efficiency of the human brain and the energy gulping of computer mainframes. They essentially showed that neuromorphic hardware \u2014 chips designed to emulate the sparse, asynchronous communication of biological brains \u2014 can directly solve the complex partial differential equations (PDEs) that underpin our understanding of the physical world and form the bedrock of scientific simulation.<\/p>\n

By translating the trusted mathematics of structural mechanics into the language of spiking neurons, the team has opened a backdoor to energy-efficient supercomputing that looks less like a processor of ones and zeroes and more like a living mind.<\/p>\n

The Problem with Simulating the World<\/p>\n

Whether forecasting a hurricane\u2019s path or testing a nuclear warhead, scientists rely on PDEs.<\/p>\n

To solve these on a computer, engineers use the Finite Element Method (FEM). They take a complex shape \u2014 say, an airplane wing \u2014 and break it down into a \u201cmesh\u201d of millions of tiny, simple geometric elements. Solving the math for these millions of elements requires massive supercomputers that guzzle electricity and generate immense heat.<\/p>\n

The huge energy expenditure is partly owed to the way computer architecture is currently designed. Traditional chips spend vast amounts of energy shuttling numbers back and forth between memory and processors. The brain, however, doesn\u2019t work that way. It keeps memory and computation together, distributed across billions of neurons.<\/p>\n

\u201cWe\u2019re just starting to have computational systems that can exhibit intelligent-like behavior. But they look nothing like the brain, and the amount of resources that they require is ridiculous, frankly,\u201d says Brad Theilman, a computational neuroscientist at Sandia.<\/p>\n

The \u201cNeuroFEM\u201d Breakthrough<\/p>\n

\"\"<\/a>Infographic by ZME Science.<\/p>\n

Theilman and his colleague James Aimone didn\u2019t try to train a neural network to guess the answer to physics problems, as many deep learning AI models do. Instead, they found a way to translate the exact mathematics of the Finite Element Method into a Spiking Neural Network (SNN).<\/p>\n

\u00d7<\/p>\n

Thank you! One more thing…<\/p>\n

Please check your inbox and confirm your subscription.<\/p>\n

They call it NeuroFEM.<\/p>\n

In their system, the mesh of the physical object is mapped onto a mesh of neurons. Instead of passing complex floating-point numbers (like 3.14159) back and forth, these neurons communicate via \u201cspikes\u201d\u2014tiny, binary pulses of electricity, and are meant to mimic biological neural spiking<\/a>.<\/p>\n

It functions like a microscopic tug-of-war. For every point in the mesh, a small population of neurons receives input and \u201cspikes\u201d to signal a value. Half the neurons push the value positive, and half push it negative. Through this rapid-fire, asynchronous communication, the network naturally flows toward a balance point. That balance point is the solution to the equation. <\/p>\n

\u201cYou can solve real physics problems with brain-like computation,\u201d Aimone says. \u201cThat\u2019s something you wouldn\u2019t expect because people\u2019s intuition goes the opposite way. And in fact, that intuition is often wrong\u201d.<\/p>\n

Silicon That Scales<\/p>\n

To prove this wasn\u2019t just a blackboard theory, the team ran their algorithm on Intel\u2019s Loihi 2<\/a>, a cutting-edge neuromorphic chip.<\/p>\n

The results were startlingly efficient. The researchers found that their algorithm exhibited \u201cclose to ideal scaling\u201d. In traditional computing, adding more processors often yields diminishing returns due to data traffic jams. As you add more, you run out of low-hanging fruit and the whole setup becomes increasingly economically unviable. But with NeuroFEM on Loihi 2, doubling the number of cores nearly halved the time required to solve the problem. <\/p>\n

Conversely, the energy cost to reach a solution was significantly lower than running the same math on a standard CPU. As the problems get larger and more complex, this energy advantage is expected to grow. <\/p>\n

From Tennis Balls to Warheads<\/p>\n

Why does a chip designed to mimic the brain excel at physics? It turns out, your brain is doing this kind of math all the time.<\/p>\n

\u201cPick any sort of motor control task \u2014 like hitting a tennis ball or swinging a bat at a baseball,\u201d Aimone explains. \u201cThese are very sophisticated computations. They are exascale-level problems that our brains are capable of doing very cheaply\u201d.<\/p>\n

The algorithm they used is actually based on a model of the brain\u2019s motor cortex. The same neural architecture evolution built to control your arm movement is mathematically perfectly suited to simulate the bending of a steel beam. That\u2019s a pretty wild thought. <\/p>\n

This has massive implications for the National Nuclear Security Administration (NNSA), which funded the work. The NNSA relies on massive simulations to maintain the nuclear deterrent without physically detonating hydrogen bombs. <\/p>\n

\u201cNeuromorphic computing may provide a way to significantly cut energy use while still delivering strong computational performance,\u201d effectively allowing for larger, faster simulations on a smaller power budget, according to the researchers.<\/p>\n

The \u201cNeuromorphic Twin\u201d<\/p>\n

Perhaps the most exciting application is the concept of the \u201cneuromorphic twin\u201d.<\/p>\n

Because these chips are low-power and process data in real-time spikes, they could be embedded directly into physical structures, like a bridge or a turbine. The chip could run a continuous simulation of the object it is embedded in and the forces that act upon it in real-time, updating instantly based on sensor data to predict structural failure before it happens.<\/p>\n

The team even demonstrated that their network could handle complex 3D shapes, such as a hollow sphere deforming under gravity, proving it can handle the messy, unstructured geometry of the real world.<\/p>\n

One of the biggest criticisms of modern AI in science is the \u201cblack box\u201d problem. We often don\u2019t know how an AI gets its answer. NeuroFEM is different.<\/p>\n

\u201cIf we\u2019ve already shown that we can import this relatively basic but fundamental applied math algorithm into neuromorphic \u2014 is there a corresponding neuromorphic formulation for even more advanced applied math techniques?\u201d Theilman asks. <\/p>\n

As development continues, the researchers are optimistic. \u201cWe have a foot in the door for understanding the scientific questions, but also we have something that solves a real problem,\u201d Theilman added.<\/p>\n

The findings appeared in the journal Nature Machine Intelligence<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"AI-generated illustration. Credit: ZME Science\/Nanobanana. When you swing a tennis racket or catch a set of keys, you…\n","protected":false},"author":2,"featured_media":296320,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[7],"tags":[353,128698,85,46,150175,141,10466],"class_list":{"0":"post-296319","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-science","8":"tag-computing","9":"tag-finite-element-analysis","10":"tag-il","11":"tag-israel","12":"tag-neuromorphic-computer","13":"tag-science","14":"tag-supercomputer"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/posts\/296319","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/comments?post=296319"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/posts\/296319\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/media\/296320"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/media?parent=296319"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/categories?post=296319"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/tags?post=296319"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}