Multi-Objective Instruction-Aware Representation Learning in Procedural Content Generation RL

TL;DR

This paper introduces MIPCGRL, a multi-objective representation learning approach that enhances controllability in instructed procedural content generation using sentence embeddings.

Contribution

MIPCGRL is a novel method that effectively trains a multi-objective embedding space for complex instructions, improving controllability in content generation tasks.

Findings

Achieves up to 13.8% improvement in controllability.

Effectively processes complex, multi-objective instructions.

Enables more expressive and flexible content generation.

Abstract

Recent advancements in generative modeling emphasize the importance of natural language as a highly expressive and accessible modality for controlling content generation. However, existing instructed reinforcement learning for procedural content generation (IPCGRL) method often struggle to leverage the expressive richness of textual input, especially under complex, multi-objective instructions, leading to limited controllability. To address this problem, we propose \textit{MIPCGRL}, a multi-objective representation learning method for instructed content generators, which incorporates sentence embeddings as conditions. MIPCGRL effectively trains a multi-objective embedding space by incorporating multi-label classification and multi-head regression networks. Experimental results show that the proposed method achieves up to a 13.8\% improvement in controllability with multi-objective instructions. The ability to process complex instructions enables more expressive and flexible content generation.