解决多模态行为克隆中的模式坍缩问题，提升机器人动作建模多样性。
EBGAN-MDN: An Energy-Based Adversarial Framework for Multi-Modal Behavior Cloning
- 用能量模型与混合密度网络结合，对抗训练增强多模态表达能力。
- 在合成数据和机器人基准上性能优于传统方法，避免模式平均现象。
- 适合需要多样动作输出的机器人控制任务，如自动驾驶、人机交互。
多模态行为克隆面临模式平均和模式坍缩的挑战，传统模型难以捕捉输入输出间的多样化映射关系。该问题在机器人应用中尤为关键，准确建模多种有效动作可保障性能与安全。本文提出EBGAN-MDN框架，融合能量基模型、混合密度网络（MDN）与对抗训练。通过改进的InfoNCE损失与能量约束的MDN损失，有效缓解上述问题。在合成数据与机器人基准测试中，该方法表现优越，验证了其在多模态学习任务中的有效性与高效性。
原文摘要 · Abstract (English)
Multi-modal behavior cloning faces significant challenges due to mode averaging and mode collapse, where traditional models fail to capture diverse input-output mappings. This problem is critical in applications like robotics, where modeling multiple valid actions ensures both performance and safety. We propose EBGAN-MDN, a framework that integrates energy-based models, Mixture Density Networks (MDNs), and adversarial training. By leveraging a modified InfoNCE loss and an energy-enforced MDN loss, EBGAN-MDN effectively addresses these challenges. Experiments on synthetic and robotic benchmarks demonstrate superior performance, establishing EBGAN-MDN as a effective and efficient solution for multi-modal learning tasks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。