 |
|
|
|
||||||||||||
 |
|||||||||||||||
|
 
University of Pennsylvania
Machine Learning and Robotics Why is it that if computers have gotten so much faster and cheaper, they have not become any better at understanding what we want them to do? Some of the tasks we take for granted such as vision and language are still too difficult for the fastest supercomputers to handle. To us, a picture may be worth a thousand words, but to a machine it's just a seemingly random jumble of numbers. How can we get machines to intelligently process this kind of information? Algorithms that mimic the way biological brains compute and learn may be the answer. I believe that in order for computers to sense and respond intelligently to our actions, they need to be endowed with the fundamental ability to adapt and learn from experience. But with data-intensive, multimodal inputs such as audio and video, many computational algorithms suffer from information overload. It is very difficult to detect the important features to attend to and to learn from. The theoretical research in my lab is focused on the problem of extracting the underlying key features from such data. My lab is also constructing artificial systems to investigate the real-world problems inherent in human-machine interactions. They provide data for testing theoretical algorithms and also illustrate potential applications.
Graduate Groups |
 |
|||||||||||||
 |
|||||||||||||||
|
home | faculty | grads | undergrads | departments | institutes | courses | talks Contact: perception@psych.upenn.edu |
|||||||||||||||
