{"id":218281,"date":"2025-07-19T02:20:43","date_gmt":"2025-07-19T07:20:43","guid":{"rendered":"https:\/\/lifeboat.com\/blog\/2025\/07\/common-feature-between-forest-fires-and-neural-networks-reveals-universal-framework"},"modified":"2025-07-19T02:20:43","modified_gmt":"2025-07-19T07:20:43","slug":"common-feature-between-forest-fires-and-neural-networks-reveals-universal-framework","status":"publish","type":"post","link":"https:\/\/lifeboat.com\/blog\/2025\/07\/common-feature-between-forest-fires-and-neural-networks-reveals-universal-framework","title":{"rendered":"Common feature between forest fires and neural networks reveals universal framework"},"content":{"rendered":"

<\/a><\/p>\n

Researchers from the University of Tokyo in collaboration with Aisin Corporation have demonstrated that universal scaling laws, which describe how the properties of a system change with size and scale, apply to deep neural networks that exhibit absorbing phase transition behavior, a phenomenon typically observed in physical systems. The discovery not only provides a framework describing deep neural networks but also helps predict their trainability or generalizability. The findings were published<\/a> in the journal Physical Review Research.<\/p>\n

In recent years, it seems no matter where we look, we come across artificial intelligence<\/a> in one form or another. The current version of the technology is powered by deep neural networks<\/a>: numerous layers of digital \u201cneurons\u201d with weighted connections between them. The network learns by modifying the weights between the \u201cneurons\u201d until it produces the correct output. However, a unified theory<\/a> describing how the signal propagates between the layers of neurons in the system has eluded scientists so far.<\/p>\n

\u201cOur research was motivated by two drivers,\u201d says Keiichi Tamai, the first author. \u201cPartially by industrial needs as brute-force tuning of these massive models takes a toll on the environment. But there was a second, deeper pursuit: the scientific understanding of the physics of intelligence itself.\u201d<\/p>\n","protected":false},"excerpt":{"rendered":"

Researchers from the University of Tokyo in collaboration with Aisin Corporation have demonstrated that universal scaling laws, which describe how the properties of a system change with size and scale, apply to deep neural networks that exhibit absorbing phase transition behavior, a phenomenon typically observed in physical systems. The discovery not only provides a framework [\u2026]<\/p>\n","protected":false},"author":427,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[219,6],"tags":[],"class_list":["post-218281","post","type-post","status-publish","format-standard","hentry","category-physics","category-robotics-ai"],"_links":{"self":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts\/218281","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/users\/427"}],"replies":[{"embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/comments?post=218281"}],"version-history":[{"count":0,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts\/218281\/revisions"}],"wp:attachment":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/media?parent=218281"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/categories?post=218281"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/tags?post=218281"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}