SCALES: Smart Contracts (on blockchain platforms) Augmented with LEarning and Semantics


Smart contracts enable execution of simple programs that carry out transactions on the blockchain. Since smart contracts are immutable, trust and transparency are preserved. However, many decentralized applications require intelligence beyond the execution of logical constructs conceptualized initially. A “break glass” scenario involving access to a patient's medical records, or coordination of internet of things devices to share data in an initially unforeseen circumstance are some motivating examples. Constructing smart contracts that act as autonomous agents with feedback, and rethinking the execution to make allowances for data ascertained at a later point in time is a novel area of research. In this project we will address the challenge of agent driven smart contracts on the blockchain with semantics, advances in machine learning, and state of the art in multi agent systems research. These smart agents will be analyzed analytically and empirically using game theory and agent-based techniques.