Distributed creation of Machine learning agents for Blockchain analysis

TL;DR

This paper introduces a blockchain-based protocol incentivizing distributed nodes to run neural architecture search algorithms, aiming to create autonomous, self-improving AI models for cryptocurrency prediction and broader applications.

Contribution

It proposes a novel blockchain protocol that encourages independent computation nodes to collaboratively optimize neural networks, enhancing automation and democratization of AI development.

Findings

Customized NAS with network morphism and Bayesian optimization achieved competitive results.

Distributed NAS can potentially surpass human-designed models with sufficient computing power.

The blockchain protocol incentivizes autonomous, self-improving AI model generation.

Abstract

Creating efficient deep neural networks involves repetitive manual optimization of the topology and the hyperparameters. This human intervention significantly inhibits the process. Recent publications propose various Neural Architecture Search (NAS) algorithms that automate this work. We have applied a customized NAS algorithm with network morphism and Bayesian optimization to the problem of cryptocurrency predictions, where it achieved results on par with our best manually designed models. This is consistent with the findings of other teams, while several known experiments suggest that given enough computing power, NAS algorithms can surpass state-of-the-art neural network models designed by humans. In this paper, we propose a blockchain network protocol that incentivises independent computing nodes to run NAS algorithms and compete in finding better neural network models for a particular task. If implemented, such network can be an autonomous and self-improving source of machine learning models, significantly boosting and democratizing the access to AI capabilities for many industries.