Li Fei-Fei

TL;DR: This paper proposed a model that maximizes mutual information between the image, the expected answer and the generated question to generate more diverse, goal-driven questions by regularizing this latent space with a second latent space that ensures clustering of similar answers.

TL;DR: Evidence is provided for a new organizational principle, in which scene processing relies on two distinct networks that split the classically defined Parahippocampal Place Area, which is involved in a much broader set of tasks involving episodic memory and navigation.

TL;DR: The authors propose a continuous relaxation of the discrete symbolic planner that directly plans on the probabilistic outputs of the symbol grounding model to disentangle the policy execution from the inter-task generalization and lead to better data efficiency.

TL;DR: This work proposes a non-intrusive vision-based system for tracking people's activity in hospitals, and demonstrates promising results for reducing hospital acquired infections.

TL;DR: This work proposes a reinforcement learning-based formulation for selecting the right examples for training a classifier from noisy web search results and uses Q-learning to learn a data labeling policy on a small labeled training dataset, and then uses this to automatically label noisy web data for new visual concepts.