Deep neural network approximation theory
AuthorsPhilipp Grohs, Dmytro Perekrestenko, Dennis Elbrächter, and Helmut Bölcskei
ReferenceIEEE Transactions on Information Theory, Jan. 2019, submitted.
AbstractDeep neural networks have become state-of-the-art technology for a wide range of practical machine learning tasks such as image classification, handwritten digit recognition, speech recognition, or game intelligence. This paper develops the fundamental limits of learning in deep neural networks by characterizing what is possible if no constraints on the learning algorithm and the amount of training data are imposed. Concretely, we consider information-theoretically optimal approximation through deep neural networks with the guiding theme being a relation between the complexity of the function (class) to be approximated and the complexity of the approximating network in terms of connectivity and memory requirements for storing the network topology and the associated quantized weights. The theory we develop educes remarkable universality properties of deep networks. Specifically, deep networks are optimal approximants for vastly different function classes such as affine systems and Gabor systems. Affine systems are generated by the affine group (scalings and translations) whereas Gabor systems are generated by the Weyl-Heisenberg group (time-shifts and modulations). This universality is afforded by a concurrent invariance property of deep networks to time-shifts, scalings, and frequency-shifts. In addition, deep networks provide exponential approximation accuracy—i.e., the approximation error decays exponentially in the number of non-zero weights in the network—of vastly different functions such as the squaring operation, multiplication, polynomials, sinusoidal functions, general smooth functions, and even one-dimensional oscillatory textures and fractal functions such as the Weierstrass function, both of which do not have any known methods achieving exponential approximation accuracy. In summary, deep neural networks provide information-theoretically optimal approximation of a very wide range of functions and function classes used in mathematical signal processing. We also show that in the approximation of sufficiently smooth functions finite-width deep networks require strictly smaller connectivity than finite-depth wide networks.
KeywordsDeep learning, neural networks, approximation theory, Kolmogorov rate-distortion theory
Download this document:
Copyright Notice: © 2019 P. Grohs, D. Perekrestenko, D. Elbrächter, and H. Bölcskei.
This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible.
This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.