[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-training-deep-neural-nets-to-run-on-low-power-ising-hardware":10,"sections":34},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":24,"tags":25,"sources":29,"feedback":33,"feedback_at":22,"cost_usd":33,"total_tokens":33},3222,"training-deep-neural-nets-to-run-on-low-power-ising-hardware","Training Deep Neural Nets to Run on Low-Power Ising Hardware","Researchers have a working method to train convolutional networks for thermodynamic Ising machines, hitting 94.9% accuracy on a standard image benchmark.","A new training method lets deep convolutional networks run on thermodynamic Ising machine hardware without rewriting how the models are trained.\n\nResearchers published a backpropagation-based algorithm that trains deep convolutional networks to run inference on Ising machine hardware — physical devices that exploit thermodynamic noise rather than conventional transistor logic. The models hit 94.9% accuracy on CIFAR-10 and 76.0% on CIFAR-100 under binary Gibbs sampling, competitive with conventional approaches. The team also derived a mathematical relationship between inference cost and accuracy, then used it to calculate optimal inference schedules. The theory was experimentally validated, not just modeled on paper.\n\nIf this scales, it matters because Ising machines consume far less power than GPU-based inference, which makes them attractive for edge deployments where energy budgets are tight. The persistent gap has been on the training side: prior work established that Ising systems could theoretically run inference, but lacked practical methods to train large models for them. This paper closes that gap with standard backprop, meaning existing deep learning tooling applies.\n\nThe result is early-stage research, not a shipping product — and Ising hardware itself remains a niche compared to the GPU supply chains that every major AI lab has already locked in.","[\"ai\",\"hardware\",\"machine-learning\",\"edge-computing\"]","2026-07-02T04:00:00.000Z","2026-07-02T05:05:48.591Z","2026-07-02T05:05:51.648Z","published",null,[],"ai",[24,26,27,28],"hardware","machine-learning","edge-computing",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2607.00170",0,{"sections":35},[36,40,45,50,55,59,64,69,74,79,84,88,93,98],{"name":37,"slug":24,"count":38,"latest_published_at":39},"AI",2590,"2026-07-16T04:00:00.000Z",{"name":41,"slug":42,"count":43,"latest_published_at":44},"Security","security",294,"2026-07-15T19:59:48.000Z",{"name":46,"slug":47,"count":48,"latest_published_at":49},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":51,"slug":52,"count":53,"latest_published_at":54},"Policy","policy",158,"2026-07-16T00:02:48.000Z",{"name":56,"slug":26,"count":57,"latest_published_at":58},"Hardware",122,"2026-07-14T19:46:26.000Z",{"name":60,"slug":61,"count":62,"latest_published_at":63},"Consumer Tech","consumer-tech",93,"2026-07-13T13:20:48.000Z",{"name":65,"slug":66,"count":67,"latest_published_at":68},"Software","software",70,"2026-07-13T19:52:25.000Z",{"name":70,"slug":71,"count":72,"latest_published_at":73},"Science","science",66,"2026-07-10T10:29:37.000Z",{"name":75,"slug":76,"count":77,"latest_published_at":78},"Dev Tools","dev-tools",59,"2026-07-07T04:00:00.000Z",{"name":80,"slug":81,"count":82,"latest_published_at":83},"Gaming","gaming",41,"2026-07-09T04:00:00.000Z",{"name":85,"slug":86,"count":82,"latest_published_at":87},"Startups","startups","2026-06-29T20:55:50.000Z",{"name":89,"slug":90,"count":91,"latest_published_at":92},"General","general",29,"2026-07-10T22:28:58.000Z",{"name":94,"slug":95,"count":96,"latest_published_at":97},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":99,"slug":100,"count":101,"latest_published_at":102},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]