Year: 2017
Type: preprint
Abstract: This paper describes a family of probabilistic architectures designed for online learning under the logarithmic loss. Rather than relying on non-linear transfer functions, our method gains representational power by the use of data conditioning. We state under general conditions a learnable capacity theorem that shows this approach can in principle learn any bounded Borel-measurable function on a compact subset of euclidean space; the result is stronger than many universality results for connectionist architectures because we provide both the model and the learning procedure for which convergence is guaranteed.
Source: arXiv (Cornell University)
Authors: Veness, Joel, Lattimore, Tor, Bhoopchand, Avishkar, Grabska-Barwinska, Agnieszka, Mattern, Christopher, Toth, Peter
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2771657877