The Robotic Baby

Before you learn how to drive, you learn how to be a baby. Actually, babies are born knowing how to learn. I forgot how complicated it was to be a baby.
Free live streaming by Ustream

Watch and listen to Lisa Meeden from Swarthmore College as she speaks about Creating Curious Robots. The talk is streamed at http://www.ustream.tv/channel/umbc-ebiquity-meeting.

Abstract: Applying machine learning to a robotics problem typically requires substantial human oversight to design the learning system, tune the parameters, define the task, determine the input and output representations, and create the training data set. In contrast, biological organisms are able to learn autonomously from unlabeled data in an open-ended fashion. Developmental robotics is an emerging field that strives to build better robots by applying insights from biological developmental processes. In this talk I will review several recent approaches from developmental robotics that use prediction to generate teaching signals. This results in a task-independent kind of learning in which the robot focuses on novel stimuli.

The streaming video site is blocked at work; I was told to watch it at home. If you watch the UMBC ebiquity weekly talks in real time (7-8 AM in LA), you can type questions in for the speaker. I like attending seminars in my jammies.

The part around minute 7, when she explains why we ignore the most familiar, was particularly worth noting.

The UMBC ebiquity blog is also worth reading if you are interested in the technological stuff "under the hood" that is shaping the way humans interact with computers.

The most recent talks are listed first at http://www.ustream.tv/channel/umbc-ebiquity-meeting. The talk before Lisa Meedan's may appear more arcane, but skip to minute 46 for a worthwhile discussion of net-neutrality and bandwidth throttling.