There has been an emerging paradigm shift from the era of “internet AI” to “embodied AI,”
where AI algorithms and agents no longer learn from datasets of images, videos or text …

More than one hundred benchmarks have been developed to test the commonsense
knowledge and commonsense reasoning abilities of artificial intelligence (AI) systems …

The ability to reason about temporal and causal events from videos lies at the core of human
intelligence. Most video reasoning benchmarks, however, focus on pattern recognition from …

Physics-informed machine learning (PIML), referring to the combination of prior knowledge
of physics, which is the high level abstraction of natural phenomenons and human …

Despite recent successes of reinforcement learning (RL), it remains a challenge for agents
to transfer learned skills to related environments. To facilitate research addressing this …

Computer vision has undergone a dramatic revolution in performance, driven in large part
through deep features trained on large-scale supervised datasets. However, much of these …

In this work, we propose a unified framework, called Visual Reasoning with Differ-entiable
Physics (VRDP), that can jointly learn visual concepts and infer physics models of objects …

Human visual perception carves a scene at its physical joints, decomposing the world into
objects, which are selectively attended, tracked and predicted as we engage our …

We study the problem of dynamic visual reasoning on raw videos. This is a challenging
problem; currently, state-of-the-art models often require dense supervision on physical …

One of the primary challenges limiting the applicability of deep learning is its susceptibility to
learning spurious correlations rather than the underlying mechanisms of the task of interest …