d7b3cef7c31b94a4a533db83d01a8882-Paper-Conference.pdf

Neural Information Processing Systems 

Latent action models (LAMs) aim to learn action-relevant changes from unlabeled videos by compressing changes between frames as latents. However, differences between video frames can be caused by controllable changes as well as exogenous noise, leading to an important concern - do latents capture the changes caused by actions or irrelevant noise?