[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-new-method-lets-ai-learn-hidden-state-systems-from-data":10,"sections":35},{"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":30,"feedback":34,"feedback_at":22,"cost_usd":34,"total_tokens":34},6369,"new-method-lets-ai-learn-hidden-state-systems-from-data","New Method Lets AI Learn Hidden-State Systems From Data","Researchers built an algorithm that learns hidden system dynamics like locks from interaction data alone, then proved a hard limit on what it can ever learn.","A new algorithm lets AI agents reverse-engineer how hidden-state systems, like a lock, actually work using nothing but a log of actions and observations.\n\nResearchers built a method for learning the full parameters of a Partially Observable Markov Decision Process, or POMDP, a standard way to model systems whose true state is invisible to the agent. Their approach borrows from spectral learning techniques such as Predictive State Representations, but pushes further: instead of just predicting outcomes, it recovers explicit transition and observation matrices using tensor decomposition, up to a mathematical similarity transform. In numerical experiments, the team found that these explicit models let an agent generate new plans for goals and reward functions it was never trained on. The method depends on a set of rankness assumptions the transition and observation matrices must satisfy to work at all.\n\nThat distinction matters because most spectral approaches to partial observability stop at prediction. A PSR can tell you what is likely to happen next, but it cannot be repurposed for a different task without retraining from scratch. Explicit transition and observation models can, in principle, plug into a planner for a brand new objective, which is closer to how engineers actually want autonomous systems to behave once deployed.\n\nThe researchers also proved a hard ceiling on how far this can go: they constructed two POMDPs that produce identical observation statistics forever, yet have different internal transition dynamics. No amount of sequential data can tell them apart. It is a tidy reminder that learning from behavior alone has a mathematical floor, not just an engineering one.","[\"ai research\",\"pomdp\",\"reinforcement learning\",\"machine learning\"]","2026-09-11T04:00:00.000Z","2026-09-11T08:57:33.974Z","2026-09-11T08:57:45.906Z","published",null,[],"ai",[26,27,28,29],"ai research","pomdp","reinforcement learning","machine learning",[31],{"name":32,"url":33},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2601.18930",0,{"sections":36},[37,40,44,48,53,58,63,66,71,75,80,85,90,95],{"name":38,"slug":24,"count":39,"latest_published_at":18},"AI",3543,{"name":41,"slug":42,"count":43,"latest_published_at":18},"Security","security",637,{"name":45,"slug":46,"count":47,"latest_published_at":18},"Policy","policy",338,{"name":49,"slug":50,"count":51,"latest_published_at":52},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":54,"slug":55,"count":56,"latest_published_at":57},"Hardware","hardware",153,"2026-09-09T15:12:32.000Z",{"name":59,"slug":60,"count":61,"latest_published_at":62},"Consumer Tech","consumer-tech",99,"2026-09-09T17:27:33.000Z",{"name":64,"slug":65,"count":61,"latest_published_at":18},"Science","science",{"name":67,"slug":68,"count":69,"latest_published_at":70},"Software","software",75,"2026-09-10T20:41:21.000Z",{"name":72,"slug":73,"count":74,"latest_published_at":18},"Dev Tools","dev-tools",70,{"name":76,"slug":77,"count":78,"latest_published_at":79},"Startups","startups",55,"2026-09-09T23:14:29.000Z",{"name":81,"slug":82,"count":83,"latest_published_at":84},"Gaming","gaming",43,"2026-09-10T12:18:06.000Z",{"name":86,"slug":87,"count":88,"latest_published_at":89},"General","general",41,"2026-09-08T01:57:23.000Z",{"name":91,"slug":92,"count":93,"latest_published_at":94},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":96,"slug":97,"count":98,"latest_published_at":99},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]