Backprop as functor. A compositional perspective on supervised learning

DOI10.5555/3470152.3470163MaRDI QIDQ6654476zbMATH OpenFDO

Authors Brendan Fong, David I. Spivak, Rémy Tuyéras

Publication date 19 December 2024

Full work available at URL https://doi.org/10.5555/3470152.3470163

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Recommendations

• Learners' languages
• Categorical foundations of gradient-based learning
• Neural nets via forward state transformation and backward loss transformation
• Reverse derivative ascent: a categorical approach to learning Boolean circuits
• General supervised learning as change propagation with delta lenses

Cited in

(11)

• Effectful semantics in 2-dimensional categories: premonoidal and Freyd bicategories
• Monoidal structures on generalized polynomial categories
• Effectful semantics in bicategories: strong, commutative, and concurrent pseudomonads
• The graphical theory of monads
• Differentiable causal computations via delayed trace (extended version)
• Categorical foundations of gradient-based learning
• A categorical model for organic chemistry
• Homotopical algebra and higher structures. Abstracts from the workshop held August 26--30, 2024
• Learners' languages
• Limits of learning dynamical systems
• Operads for complex system design specification, analysis and synthesis

This page was built for publication: Backprop as functor. A compositional perspective on supervised learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6654476)