Discovering Continuous-Time Memory-Based Symbolic Policies using Genetic Programming

2024-06-04

AI Summary

To address the limited interpretability of AI controllers in partially observable and dynamic environments, this paper proposes a symbolic white-box control method integrated with a continuous-time memory mechanism: a latent-variable memory module—modeled by ordinary differential equations—is embedded within a symbolic policy and automatically evolved via genetic programming to yield controllers that are both mathematically tractable and dynamically adaptive. This work establishes the first synergistic modeling framework combining continuous-time memory with symbolic regression, overcoming the expressive limitations of conventional memoryless symbolic approaches in non-Markovian settings. Experimental results demonstrate that the proposed memory-augmented symbolic controller achieves performance on par with black-box models across multiple control tasks, significantly outperforming memoryless baselines, while simultaneously ensuring high transparency, strong robustness, and formal verifiability.

Technology Category

• Cognitive Modeling & Cognitive Systems: Symbolic Representations
• Machine Learning: Neuro-Symbolic Learning
• Intelligent Robots: Behavior Learning & Control

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Artificial intelligence techniques are increasingly being applied to solve control problems, but often rely on black-box methods without transparent output generation. To improve the interpretability and transparency in control systems, models can be defined as white-box symbolic policies described by mathematical expressions. For better performance in partially observable and volatile environments, the symbolic policies are extended with memory represented by continuous-time latent variables, governed by differential equations. Genetic programming is used for optimisation, resulting in interpretable policies consisting of symbolic expressions. Our results show that symbolic policies with memory compare with black-box policies on a variety of control tasks. Furthermore, the benefit of the memory in symbolic policies is demonstrated on experiments where memory-less policies fall short. Overall, we present a method for evolving high-performing symbolic policies that offer interpretability and transparency, which lacks in black-box models.

Problem

• Develop interpretable symbolic policies for control systems
• Extend policies with memory for volatile environments
• Optimize policies using genetic programming for transparency

Innovation

• Genetic programming optimizes symbolic policies
• Continuous-time memory enhances symbolic policies
• White-box models ensure interpretable control systems