Besides standard latent subsymbolic variables, our model exploits a probabilistic logicprogram todefine afurther structured representation, which is used forlogical reasoning.

Cutting-plane methods have enabled remarkable successes in integer programming over the last few decades.

Self-supervised representation learning approaches haverecently surpassed their supervised learning counterparts on downstream tasks like object detection andimage classification.

These techniques mainly fall in one of two categories: value-based methods [e.g.,

Unsupervised domain adaptation (UDA) plays a crucial role in addressing distribution shifts in machine learning. In this work, we improve the theoretical foundations of UDA proposed in Acuna et al. (2021) by refining their

Inthis paper,we first investigate twokinds oftrivial solutions in the compositional generation process, and demonstrate their source isvanishing gradients onthemask.

However, both L2 and BL have two deficiencies. First, the noise in the annotation process is not considered in a principled way. L2 and BL make an assumption about per-pixel i.i.d.

The essence of object detection lies in training a detection backbone to extract visual features from images and a detection head to recognize and locate objects based on the visual features.