The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence

TL;DR

This paper advocates for a hybrid AI approach combining knowledge-driven reasoning with cognitive models to develop more robust and versatile artificial intelligence systems, contrasting with current trends focused on scale and data.

Contribution

It introduces a novel hybrid framework integrating cognitive models and reasoning, aiming to enhance AI robustness beyond current scale-focused methods.

Findings

Proposes a reasoning-based hybrid AI approach.

Highlights limitations of scale-centric AI methods.

Suggests cognitive models as a foundation for robustness.

Abstract

Recent research in artificial intelligence and machine learning has largely emphasized general-purpose learning and ever-larger training sets and more and more compute. In contrast, I propose a hybrid, knowledge-driven, reasoning-based approach, centered around cognitive models, that could provide the substrate for a richer, more robust AI than is currently possible.