Colloquium: David Anderson

Abstract: Great strides have been made in machine learning, especially in image, speech, and natural language understanding due to the large datasets and the application of tremendous resources. However, for many problems, there is a lack of good data and labels and other characteristics needed for traditional machine learning. We have been exploring methods of learning from unlabeled and/or weakly labeled data by “learning normal” and how to characterize deviations from normal.

Bio: Dr. David Anderson has been a professor at Georgia Tech for nearly 20 years. Prior to that, he received his BS and MS from Brigham Young University and PhD from Georgia Tech in Electrical and Computer Engineering. He has received numerous awards including the Presidential Early Career Award for Scientists and Engineers. His core research is in signal processing but his research has ranged over many collaborative areas including IC design, processor architectures, and a variety of applications of signal processing and machine learning to audio and image processing problems.