Anomaly detection is important for data cleaning, cybersecurity, and robust AI systems. This talk will review recent work from our group on (a) benchmarking existing algorithms, (b) developing a theoretical understanding of their behavior, (c) explaining anomalies/"alarms" to a data analyst, and (d) interactively re-evaluating possible anomalies in response to the analyst's feedback. The talk then describes two applications: (a) detecting and diagnosing sensor failures in weather networks, and (b) detecting open categories in supervised learning.

For more information, see https://www.microsoft.com/en-us/research/video/anomaly-detection-algorithms-explanations-applications/

Please take the opportunity to connect with your friends and family and share this video with them if you find it useful.