[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-a-reinforcement-learning-fix-for-wasted-space-in-circular-packing":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},9077,"a-reinforcement-learning-fix-for-wasted-space-in-circular-packing","A Reinforcement Learning Fix for Wasted Space in Circular Packing","A new reinforcement learning system finds and repairs the leftover gaps that trip up material packing software, boosting usable space by under one percent.","A new AI system squeezes a little more material out of circular packing jobs by learning where packing algorithms get stuck.\n\nResearchers built GeoNest, a framework that pairs a reinforcement learning policy with existing geometric packing solvers to tackle the two-dimensional irregular knapsack problem: fitting oddly shaped pieces into a circular container with minimal waste. Standard solvers pack tightly at first but tend to scatter leftover space into small pockets too tiny for any remaining piece, a late-stage bottleneck. GeoNest's graph policy diagnoses which already-placed pieces are blocking a fit, then selects a bounded repair job for the underlying solver to execute. The team tested it on CircleNest-Bench, a new benchmark of 2,391 instances pulled from four contour sources, including one held-out industrial CAD dataset.\n\nMaterial utilization is a real cost line in manufacturing, since cutting sheet metal, fabric, or composites more efficiently means less scrap per job. GeoNest improved average utilization by about 0.9% on the main test sets and 0.6% on the industrial set, under the same time budget as the best standalone solver it was tested against.\n\nThose are modest numbers, but at manufacturing scale, fractions of a percent compound across thousands of cuts. It's a reminder that AI's most useful manufacturing wins right now are not flashy generative demos but incremental fixes to decades-old optimization problems.","[\"ai\",\"optimization\",\"manufacturing\",\"reinforcement-learning\"]","2026-10-01T04:00:00.000Z","2026-10-01T18:45:20.459Z","2026-10-01T18:45:21.813Z","published",null,[],"ai",[24,26,27,28],"optimization","manufacturing","reinforcement-learning",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.38863",0,{"sections":35},[36,39,43,48,53,58,62,67,72,76,81,86,91,96],{"name":37,"slug":24,"count":38,"latest_published_at":18},"AI",5496,{"name":40,"slug":41,"count":42,"latest_published_at":18},"Security","security",809,{"name":44,"slug":45,"count":46,"latest_published_at":47},"Policy","policy",429,"2026-10-01T02:26:17.000Z",{"name":49,"slug":50,"count":51,"latest_published_at":52},"Deals","deals",298,"2026-09-30T21:00:26.000Z",{"name":54,"slug":55,"count":56,"latest_published_at":57},"Hardware","hardware",196,"2026-09-30T13:00:00.000Z",{"name":59,"slug":60,"count":61,"latest_published_at":18},"Science","science",162,{"name":63,"slug":64,"count":65,"latest_published_at":66},"Consumer Tech","consumer-tech",149,"2026-09-30T22:57:11.000Z",{"name":68,"slug":69,"count":70,"latest_published_at":71},"Dev Tools","dev-tools",93,"2026-10-01T02:30:48.000Z",{"name":73,"slug":74,"count":70,"latest_published_at":75},"Software","software","2026-09-30T21:41:11.000Z",{"name":77,"slug":78,"count":79,"latest_published_at":80},"Startups","startups",84,"2026-09-30T20:39:09.000Z",{"name":82,"slug":83,"count":84,"latest_published_at":85},"Gaming","gaming",51,"2026-09-30T16:24:30.000Z",{"name":87,"slug":88,"count":89,"latest_published_at":90},"General","general",50,"2026-09-30T21:37:54.000Z",{"name":92,"slug":93,"count":94,"latest_published_at":95},"Reviews","reviews",31,"2026-09-28T14:31:34.000Z",{"name":97,"slug":98,"count":99,"latest_published_at":100},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]