Multi-Objective Instruction-Aware Representation Learning in Procedural Content Generation RL
This addresses a domain-specific problem for researchers and practitioners in procedural content generation, offering an incremental improvement in leveraging textual instructions.
The paper tackles the problem of limited controllability in instructed reinforcement learning for procedural content generation when handling complex, multi-objective textual instructions, achieving up to a 13.8% improvement in controllability.
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.