120分钟与一台笔记本：通过无监督探索和离线强化学习实现极简图像目标导航
120 Minutes and a Laptop: Minimalist Image-goal Navigation via Unsupervised Exploration and Offline RL
AI总结：
本文提出MINav方法，通过无监督数据收集与 hindsight 目标重标记，实现高效图像目标导航，在120分钟内使用消费级笔记本完成训练与部署，无需人工干预。
摘要：
图像目标视觉导航的主流范式通常假设可以访问大规模数据集、大量预训练和显著计算资源。在本文中，我们挑战这一假设。我们展示出可以在不到120分钟内收集数据集、训练领域内策略并部署到现实世界中：(1) 在120分钟内完成；(2) 在消费级笔记本上；(3) 无需任何人工干预。我们的方法MINav将图像目标导航 formulation 为一个离线目标条件强化学习问题，结合无监督数据收集与 hindsight 目标重标记和离线策略学习。在模拟和现实世界中的实验表明，MINav提高了探索效率，在目标环境中优于零样本导航基线，并且随着数据集大小的增加而表现良好。这些结果表明，通过高计算效率可以实现有效的现实世界机器人学习，降低了快速策略原型设计和部署的门槛。
英文摘要：
The prevailing paradigm for image-goal visual navigation often assumes access to large-scale datasets, substantial pretraining, and significant computational resources. In this work, we challenge this assumption. We show that we can collect a dataset, train an in-domain policy, and deploy it to the real world (1) in less than 120 minutes, (2) on a consumer laptop, (3) without any human intervention. Our method, MINav, formulates image-goal navigation as an offline goal-conditioned reinforcement learning problem, combining unsupervised data collection with hindsight goal relabeling and offline policy learning. Experiments in simulation and the real world show that MINav improves exploration efficiency, outperforms zero-shot navigation baselines in target environments, and scales favorably with dataset size. These results suggest that effective real-world robotic learning can be achieved with high computational efficiency, lowering the barrier to rapid policy prototyping and deployment.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
