{"id":77,"date":"2025-08-26T20:58:59","date_gmt":"2025-08-26T20:58:59","guid":{"rendered":"https:\/\/biconnect.psicologia.unimib.it\/?page_id=77"},"modified":"2025-09-03T17:25:53","modified_gmt":"2025-09-03T17:25:53","slug":"sac","status":"publish","type":"page","link":"https:\/\/biconnect.psicologia.unimib.it\/index.php\/sac\/","title":{"rendered":"Artificial Social Childhood (ASC)"},"content":{"rendered":"\n<h2 class=\"wp-block-heading\">Overview<\/h2>\n\n\n\n<p><strong>Artificial Social Childhood (ASC)<\/strong> creates a <strong>generative, socially rich environment<\/strong> for AI\u2014an <em>artificial childhood<\/em>\u2014to tackle a known limitation: today\u2019s LLM\/RL systems struggle with consistent <strong>causal reasoning<\/strong>, counterfactual testing, and extracting causal structure from static datasets. In ASC, agents learn in <strong>dynamic, social<\/strong> settings that mirror how children acquire knowledge: <em>active, social, ecological<\/em>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">What this grant builds<\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>A dynamic simulator (\u201csocial gym\u201d)<\/strong> grounded in an <strong>object-oriented POMDP<\/strong> where an \u201cAI child\u201d explores, acts, and asks questions to reduce uncertainty\u2014the <strong>\u201cAI of Why.\u201d<\/strong><\/li>\n\n\n\n<li><strong>Guidance &amp; explanations from an \u201cAI adult,\u201d<\/strong> the <strong>\u201cAI of Because,\u201d<\/strong> which synthesizes knowledge and provides causal feedback in digestible formats.<\/li>\n\n\n\n<li><strong>Causal learning modules<\/strong> (e.g., a <strong>causal graph<\/strong>), <strong>meta-cognitive control<\/strong> via ReAct-style planning, and <strong>adaptive, multi-source curiosity<\/strong> to drive informative exploration.<\/li>\n\n\n\n<li><strong>Curriculum of social subtasks<\/strong> to evaluate gains in causal inference, question-asking, perspective-taking, and intention recognition.<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">Why it matters<\/h2>\n\n\n\n<p>Robust decision-making depends on <strong>causal knowledge<\/strong>. By training AI in a social, interactive childhood rather than on static corpora, ASC aims to produce agents that generalize better, ask smarter questions, and reason about <strong>what would happen if\u2026<\/strong>\u2014key abilities current systems lack.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Methods at a glance<\/h2>\n\n\n\n<p>Environment \u2192 <strong>Object-Oriented POMDP<\/strong>;<br>Learner \u2192 <strong>AI Child<\/strong> (active exploration &amp; question-asking);<br>Teacher \u2192 <strong>AI Adult<\/strong> (knowledge synthesis &amp; causal explanations);<br>Control \u2192 <strong>ReAct-based meta-cognition<\/strong> for planning\/actions;<br>Motivation \u2192 <strong>Adaptive curiosity<\/strong> from multiple sources.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Programme &amp; team<\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Funder:<\/strong> <a href=\"https:\/\/www.volkswagenstiftung.de\/en\" target=\"_blank\" rel=\"noopener\">Volkswagen Foundation (VolkswagenStiftung)<\/a><\/li>\n\n\n\n<li><strong>Leads:<\/strong> <strong>Prof. Dr. Dr. Azzurra Ruggeri<\/strong> (Central European University &amp; Technical University of Munich) and <strong>Prof. Dr. Dimitri Ognibene<\/strong> (University of Milano-Bicocca)<\/li>\n<\/ul>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Overview Artificial Social Childhood (ASC) creates a generative, socially rich environment for AI\u2014an artificial childhood\u2014to tackle a known limitation: today\u2019s LLM\/RL systems struggle with consistent causal reasoning, counterfactual testing, and extracting causal structure from static datasets. In ASC, agents learn in dynamic, social settings that mirror how children acquire knowledge: active, social, ecological. What this &#8230; <a title=\"Artificial Social Childhood (ASC)\" class=\"read-more\" href=\"https:\/\/biconnect.psicologia.unimib.it\/index.php\/sac\/\" aria-label=\"Read more about Artificial Social Childhood (ASC)\">Read more<\/a><\/p>\n","protected":false},"author":3,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_acf_changed":false,"footnotes":""},"class_list":["post-77","page","type-page","status-publish"],"acf":[],"_links":{"self":[{"href":"https:\/\/biconnect.psicologia.unimib.it\/index.php\/wp-json\/wp\/v2\/pages\/77","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/biconnect.psicologia.unimib.it\/index.php\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/biconnect.psicologia.unimib.it\/index.php\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/biconnect.psicologia.unimib.it\/index.php\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/biconnect.psicologia.unimib.it\/index.php\/wp-json\/wp\/v2\/comments?post=77"}],"version-history":[{"count":3,"href":"https:\/\/biconnect.psicologia.unimib.it\/index.php\/wp-json\/wp\/v2\/pages\/77\/revisions"}],"predecessor-version":[{"id":694,"href":"https:\/\/biconnect.psicologia.unimib.it\/index.php\/wp-json\/wp\/v2\/pages\/77\/revisions\/694"}],"wp:attachment":[{"href":"https:\/\/biconnect.psicologia.unimib.it\/index.php\/wp-json\/wp\/v2\/media?parent=77"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}