𝔖 Bobbio Scriptorium
✦   LIBER   ✦

Representations and rates of approximation of real-valued Boolean functions by neural networks

✍ Scribed by V. Kůrková; P. Savický; K. Hlaváčková

Publisher
Elsevier Science
Year
1998
Tongue
English
Weight
138 KB
Volume
11
Category
Article
ISSN
0893-6080

No coin nor oath required. For personal study only.

✦ Synopsis

We give upper bounds on rates of approximation of real-valued functions of d Boolean variables by one-hidden-layer perceptron networks. Our bounds are of the form c/ n p where c depends on certain norms of the function being approximated and n is the number of hidden units. We describe sets of functions where these norms grow either polynomially or exponentially with d.

📜 SIMILAR VOLUMES

Neural network approximation of continuo
Neural network approximation of continuous functionals and continuous functions on compactifications
✍ M.B. Stinchcombe 📂 Article 📅 1999 🏛 Elsevier Science 🌐 English ⚖ 164 KB

This article characterizes the set of activation functions, bounded or unbounded, that allow feedforward network approximation of the continuous functions on the classic two-point compactification of R 1 . The characterization fails when the set of targets are continuous functions on the classic com

On Representations of Matrix Valued Neva
On Representations of Matrix Valued Nevanlinna Functions by u - Resolvents
✍ Michael Kaltenbäck; Harald Woracek 📂 Article 📅 2010 🏛 John Wiley and Sons 🌐 English ⚖ 662 KB

We show that every matrix valued generalized Nevanlinna function can be represented w a u-resolvent of a certain selfadjoint relation acting in a Pontryagin space. The negative index of this Pontryagin space may be larger than the number of negative squares of the given function. The minimal index