Welcome to the Technocracy A.I. Abstract Series for Published Scientific Work in the A.I. and Artificial General Intelligence field.  

Todays paper is titled:  Zeta Distribution and Transfer Learning Problem.

Authored By Eray Özkural.
 
Abstract: We explore the relations between the zeta distribution and algorithmic information theory via a new model of the transfer learning problem. The program distribution is approximated by a zeta distribution with parameter near 1. We model the training sequence as a stochastic process. We analyze the upper temporal bound for learning a training sequence and its entropy rates, assuming an oracle for the transfer learning problem. We argue from empirical evidence that power-law models are suitable for natural processes. Four sequence models are proposed. Random typing model is like no-free lunch where transfer learning does not work. Zeta process independently samples programs from the zeta distribution. A model of common sub-programs inspired by genetics uses a database of sub-programs. An evolutionary zeta process samples mutations from Zeta distribution. The analysis of stochastic processes inspired by evolution suggest that AI may be feasible in nature, countering no-free lunch sort of arguments.

As always thank you for listening to the Technocracy Abstract Series and a special thank you for our sponsors the Foundation, Transhumanity.net and the AGI Laboratory.

http://www.thetechnocracy.com/2018/09/24/ai-abstract-series-e10-zeta-distribution-and-transfer-learning-problem/

AI Abstract Series E10: Zeta Distribution and Transfer Learning Problem.

THE TECHNOCRACY
FROM THE MACHINE'S POINT OF VIEW…. NEWS FOR EARLY ADOPTERS, TECHNOLOGISTS, FUTURISTS, DATA SCIENTISTS, SOFTWARE ENGINEERS, TECHNOPHILES, TECHNOCRATS AND ANYONE INTERESTED IN THE NEWS.
HOME
 
Answer the most important question: What are the most important trends and the latest news?

all the news that matters, according to the machine…

A news podcast focused on news based on the analysis of Big Data by Artificial Intelligence.  A show designed to help you make sense of the sea of data rising all around us from strict analysis to future trends, technical miles stones and more.  The aesthetic of the show is framed in irreverent technical sarcasm, by experts for experts, including: early adopters, technologists, futurists, data scientists, software engineers, technophiles and technocrats alike.

AI Abstract Series E10: Zeta Distribution and Transfer Learning Problem.

The Technocracy