RA: A machine based rational agent, Part 1

TL;DR

This paper introduces RA, a software system combining machine learning and formal reasoning to discover underlying laws from data, emphasizing its initial design and conjecture construction capabilities.

Contribution

It presents the initial design of RA, integrating machine learning with formal reasoning to enable autonomous law discovery and conjecture generation.

Findings

RA can generate conjectures from empirical data

RA constructs proofs for its conjectures

Initial design strategies for RA are outlined

Abstract

RA is a software package that couples machine learning with formal reasoning in an attempt to find the laws that generate the empirical data that it has been given access to. A brief outline of RA in its initial stage of development is presented. Particular emphasis is given to current design strategies that aim to endow RA with the ability to construct its own conjectures of which it constructs proofs.