Predicting Cable Dynamics with Physical Attention Bias

Key points

• Attention over all pairs of cable segments can represent contact between parts of the cable that are far apart along its length, but attention has no notion of geometry.1
• A cable has two pairwise distances, which agree only while it is straight: the arc-length distance along the cable, which governs elastic forces, and the Euclidean distance in space, which governs contact.1
• A physical bias improves prediction on unseen cables.1
• The gain is largest when attention is the only mechanism that connects distant segments: there, the arc-length bias reduces prediction error by 15% and more than halves the drift in segment length.1

Sources (2)

• [1]Predicting Cable Dynamics with Physical Attention Bias
  Hugging Face Daily Papers · Oct 8, 12:00 AM

  Learned simulators for deformable linear objects (DLOs) such as cables have to predict the motion of cables they were not trained on and stay stable over long rollouts.

  Attention over all pairs of cable segments can represent contact between parts of the cable that are far apart along its length, but attention has no notion of geometry.

• [2]Predicting Cable Dynamics with Physical Attention Bias
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 01:51 PM · same content

Extractive summary: sentences quoted from the sources.