Event Series Brown Bag Talks

Brown Bag Talk: Jane Li & Alan Zhou

Krieger 111 @ 3400 N Charles St Baltimore, MD, United States

RNNs too expressive to properly learn local phonological patterns? Although recurrent neural networks with long short-term memory mechanisms (LSTMs) can learn many existing phonological patterns, they vastly overgenerate. This raises questions about whether LSTMs are suitable models for understanding how humans learn phonology, similar to phonological theories that overgenerate. In this work, we examine whether […]

Event Series Brown Bag Talks

Brown Bag Talk: Jieneng Chen

111 Krieger Hall 3400 N Charles St, Baltimore, MD, United States

Generative (Mental) World Explorer Abstract: Understanding, navigating, and exploring the 3D physical real world has long been a central challenge in the development of artificial intelligence. In this work, we take a step toward this goal by introducing GenEx, a system capable of planning complex embodied world exploration, guided by its generative imagination that forms priors […]

Event Series Brown Bag Talks

Brown Bag Talk: Manasi Malik

Krieger 111 @ 3400 N Charles St Baltimore, MD, United States

This talk is rescheduled from April 18 to April 25. Neural computations underlying human social evaluations from visual stimuli Abstract: Humans easily make social evaluations from visual scenes, but the computational mechanisms in the brain that support this ability remain unknown. Here, we test two hypotheses raised by prior work: one proposes that people recognize […]

Event Series Brown Bag Talks

Brown Bag Talk: Ananya Passi

Krieger 111 @ 3400 N Charles St Baltimore, MD, United States

Rapid unsupervised alignment with the natural image manifold  Abstract:There is a stark contrast between the nature of feature learning in biological and artificial vision. While brains learn without explicit supervision and with little data, deep neural networks require supervised feedback and massive training sets. Here we show that a surprisingly simple unsupervised learning algorithm can […]