Enhancing Agent Communication and Learning through Action and Language

2023/08/18 by Olivier Sigaud, Caselles-Dupré, Hugo, Sigaud, Olivier +2
Computer Science · #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Multi-Agent Systems and Negotiation #Natural Language Processing Techniques #Speech and dialogue systems

paper · pdf · doi:10.48550/arxiv.2308.10842

openalex publication_date 2023/08/18 · openalex created_date 2023/08/23 · openalex updated_date 2026/07/28

Abstract

We introduce a novel category of GC-agents capable of functioning as both teachers and learners. Leveraging action-based demonstrations and language-based instructions, these agents enhance communication efficiency. We investigate the incorporation of pedagogy and pragmatism, essential elements in human communication and goal achievement, enhancing the agents' teaching and learning capabilities. Furthermore, we explore the impact of combining communication modes (action and language) on learning outcomes, highlighting the benefits of a multi-modal approach.

Related

• Characterizing tradeoffs between teaching via language and demonstrations in multi-agent systems
• Natural Language Communication with a Teachable Agent
• Listen, Interact and Talk: Learning to Speak via Interaction
• Towards Language-Augmented Multi-Agent Deep Reinforcement Learning
• Towards Multi-Agent Communication-Based Language Learning