Roberto Cipolla

Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics

2017 · arXiv (Cornell University)

Numerous deep learning applications benefit from multi-task learning with multiple regression and classification objectives. In this paper we make the observation that the performance of such systems is strongly dependent on the relative weighting between …

Contrastive Unsupervised Learning of World Model with Invariant Causal Features

2022 · arXiv (Cornell University)

In this paper we present a world model, which learns causal features using the invariance principle. In particular, we use contrastive unsupervised learning to learn the invariant causal features, which enforces invariance across augmentations of …