Zobrazeno 1 - 10
of 14
pro vyhledávání: '"Gavranovic, Bruno"'
We propose a categorical semantics for machine learning algorithms in terms of lenses, parametric maps, and reverse derivative categories. This foundation provides a powerful explanatory and unifying framework: it encompasses a variety of gradient de
Externí odkaz:
http://arxiv.org/abs/2404.00408
Publikováno v:
Electronic Notes in Theoretical Informatics and Computer Science, Volume 4 - Proceedings of MFPS XL (December 11, 2024) entics:14638
Categories of lenses/optics and Dialectica categories are both comprised of bidirectional morphisms of basically the same form. In this work we show how they can be considered a special case of an overarching fibrational construction, generalizing Ho
Externí odkaz:
http://arxiv.org/abs/2403.16388
Autor:
Gavranović, Bruno
Deep learning, despite its remarkable achievements, is still a young field. Like the early stages of many scientific disciplines, it is marked by the discovery of new phenomena, ad-hoc design decisions, and the lack of a uniform and compositional mat
Externí odkaz:
http://arxiv.org/abs/2403.13001
Autor:
Gavranović, Bruno, Lessard, Paul, Dudzik, Andrew, von Glehn, Tamara, Araújo, João G. M., Veličković, Petar
We present our position on the elusive quest for a general-purpose framework for specifying and studying deep learning architectures. Our opinion is that the key attempts made so far lack a coherent bridge between specifying constraints which models
Externí odkaz:
http://arxiv.org/abs/2402.15332
Autor:
Gavranović, Bruno, Villani, Mattia
We define the bicategory of Graph Convolutional Neural Networks $\mathbf{GCNN}_n$ for an arbitrary graph with $n$ nodes. We show it can be factored through the already existing categorical constructions for deep learning called $\mathbf{Para}$ and $\
Externí odkaz:
http://arxiv.org/abs/2212.00542
Autor:
Gavranović, Bruno
Optics and lenses are abstract categorical gadgets that model systems with bidirectional data flow. In this paper we observe that the denotational definition of optics - identifying two optics as equivalent by observing their behaviour from the outsi
Externí odkaz:
http://arxiv.org/abs/2209.09351
Autor:
Capucci, Matteo, Gavranović, Bruno
Actions of monoidal categories on categories, also known as actegories, have been familiar to category theorists for a long time, and yet a comprehensive overview of this topic seems to be missing from the literature. Recently, actegories have been i
Externí odkaz:
http://arxiv.org/abs/2203.16351
Autor:
Braithwaite, Dylan, Capucci, Matteo, Gavranović, Bruno, Hedges, Jules, Rischel, Eigil Fjeldgren
Lenses, optics and dependent lenses (or equivalently morphisms of containers, or equivalently natural transformations of polynomial functors) are all widely used in applied category theory as models of bidirectional processes. From the definition of
Externí odkaz:
http://arxiv.org/abs/2112.11145
Over the past two decades machine learning has permeated almost every realm of technology. At the same time, many researchers have begun using category theory as a unifying language, facilitating communication between different scientific disciplines
Externí odkaz:
http://arxiv.org/abs/2106.07032
Publikováno v:
EPTCS 372, 2022, pp. 235-248
We propose a categorical framework for processes which interact bidirectionally with both an environment and a 'controller'. Examples include open learners, in which the controller is an optimiser such as gradient descent, and an approach to composit
Externí odkaz:
http://arxiv.org/abs/2105.06332