Warren Center Distinguished Lecture: Sebastian Seung
November 12, 2015 | Wu & Chen Auditorium, 101 Levine Hall
Thursday, November 12th, 2015
3:00 pm in Wu & Chen Auditorium, 101 Levine Hall
Title: The retinal connectome: from mapping to explaining brain function
Abstract: EyeWire is the world’s first game to map the brain. Anyone, anywhere in the world can help reconstruct neurons through an activity similar to a 3D coloring book. Since its launch at the end of 2012, EyeWire has registered over 200,000 players from 130 countries. EyeWire has helped draw a new wiring diagram for a neural circuit in the retina. The discovery was published by Nature magazine in 2014 with 2,183 EyeWirers as coauthors, and represents important progress towards answering the 50-year-old question of how the retina computes the direction of a moving stimulus. Over the past year, EyeWire has reconstructed a large unbiased sample of retinal neurons, aided by a Korea Telecom marketing campaign called “Countdown to Neuropia.” The reconstructions are being used to create a complete classification of retinal projection neurons into cell types, a problem that has remained unsolved for over a century. EyeWire showcases a number of advanced technologies used for brain mapping. The high resolution images come from 3D electron microscopy. When players color neurons, they interact with artificial intelligence based on “deep learning.” I will discuss research aimed at further improving these technologies, which is crucial for increasing the throughput of connectomics.
Bio: Sebastian Seung is Professor at the Princeton Neuroscience Institute and Department of Computer Science. Seung has done influential research in both computer science and neuroscience. Over the past decade, he helped pioneer the new field of connectomics, developing machine learning and social computing technologies for reconstructing neural circuits from high resolution brain images. His lab created EyeWire, a site that has recruited over 150,000 players from 130 countries to a game to map neural connections.