Universal Memory Architectures for Autonomous Machines

TL;DR

This paper introduces a self-organizing memory architecture for autonomous agents that efficiently learns and represents perceptual experiences, supporting goal-directed problem solving without prior environmental knowledge.

Contribution

It presents a novel, scalable memory architecture that combines symbolic and geometric representations, enabling autonomous learning and state space recovery.

Findings

Supports autonomous learning with quadratic space and time complexity

Provides minimal and homotopy-recovering internal representations

Utilizes duality between symbolic sets and cubical complexes

Abstract

We propose a self-organizing memory architecture for perceptual experience, capable of supporting autonomous learning and goal-directed problem solving in the absence of any prior information about the agent's environment. The architecture is simple enough to ensure (1) a quadratic bound (in the number of available sensors) on space requirements, and (2) a quadratic bound on the time-complexity of the update-execute cycle. At the same time, it is sufficiently complex to provide the agent with an internal representation which is (3) minimal among all representations of its class which account for every sensory equivalence class subject to the agent's belief state; (4) capable, in principle, of recovering the homotopy type of the system's state space; (5) learnable with arbitrary precision through a random application of the available actions. The provable properties of an effectively trained memory structure exploit a duality between weak poc sets -- a symbolic (discrete) representation of subset nesting relations -- and non-positively curved cubical complexes, whose rich convexity theory underlies the planning cycle of the proposed architecture.