Activation functions provide the non-linearity to deep neural networks, which are crucial for the optimization and performance improvement. In this paper, we propose a learnable continuous piece-wise ...

In the space of continuous functions of a real variable, the set of nowhere differentiable functions has long been known to be topologically "generic". In this paper it is shown further that in a ...

This is an experiment with continuous function optimization, a topic in the field of Nature-Inspired Optimization. This experiment tests genetic mutation algorithms on randomly generated individuals, ...

An algorithm is described for machine computation of the height of a ray originating at or near the earth's surface, when the initial elevation angle and the ray-path length (as measured, for example, ...

In discussing the behaviour of a simple queueing process for continuous time, it is shown that results, previously obtained in a rigorous way by Lederman and Reuter only by using a complicated ...