In this paper, we study the problem of unsupervised object segmentation from single images.

Training (IIT) which we call Strict IIT (SIIT). SIIT models maintain Tracr's original circuit while being more realistic.

However, most methods rely heavily on music domain knowledge.

In this section, we provide further details for our data modeling. We note the difficulties of appropriately modeling the terminal variable which is a binary variable compared to the rest of the dimensions which are continuous for the environments we investigate. This is particularly challenging for "expert" datasets where early termination is rare. An immediate advantage of sampling data from a generative model is compression. As we discuss in Appendix B.3, sampling is fast ER provides high levels of dataset compression without sacrificing downstream performance in offline reinforcement learning.