Contemporary neural networks still fall short of human-level generalization, which extends
far beyond our direct experiences. In this paper, we argue that the underlying cause for this …

Both Minsky's" society of mind" and Schmidhuber's" learning to think" inspire diverse
societies of large multimodal neural networks (NNs) that solve problems by interviewing …

Keypoint-based representation has proven advantageous in various visual and robotic
tasks. However, the existing 2D and 3D methods for detecting keypoints mainly rely on …

Current state-of-the-art object-centric models use slots and attention-based routing for
binding. However, this class of models has several conceptual limitations: the number of …

Human visual reasoning is characterized by an ability to identify abstract patterns from only
a small number of examples, and to systematically generalize those patterns to novel inputs …

The binding problem is one of the fundamental challenges that prevent the artificial neural
network (ANNs) from a compositional understanding of the world like human perception …

We present VAEL, a neuro-symbolic generative model integrating variational autoencoders
(VAE) with the reasoning capabilities of probabilistic logic (L) programming. Besides …

Visual saliency has emerged as a major visualization tool for interpreting deep
reinforcement learning (RL) agents. However, much of the existing research uses it as an …

We explore a feed-forward approach for decomposing a video into layers, where each layer
contains an object of interest along with its associated shadows, reflections, and other visual …

We propose a new representation of visual data that disentangles object position from
appearance. Our method, termed Deep Latent Particles (DLP), decomposes the visual input …