Not All Actions Contribute Equally: Learning Precise Progress Estimation from Imprecise Demonstrations

Progress estimators play an essential role in providing dense reward signals for reinforcement learning in robotic manipulation, but accurately assessing task progress can be much challenging. Existing methods heavily rely on expert data and naively assume that all actions within a demonstration contribute to task success. This assumption often breaks down in real-world scenarios, especially in high-precision manipulation, where subtle geometric errors can lead to task failure. Therefore, these methods can erroneously interpret human mistakes as meaningful progress, further corrupting…

Not summarised — no summariser is configured, so the text above is the source's own excerpt.