Anomaly Detection: Algorithms, Explanations, Applications

90,397
0
Published 2018-04-05
Anomaly detection is important for data cleaning, cybersecurity, and robust AI systems. This talk will review recent work in our group on (a) benchmarking existing algorithms, (b) developing a theoretical understanding of their behavior, (c) explaining anomaly "alarms" to a data analyst, and (d) interactively re-ranking candidate anomalies in response to analyst feedback. Then the talk will describe two applications: (a) detecting and diagnosing sensor failures in weather networks and (b) open category detection in supervised learning.

See more at www.microsoft.com/en-us/research/video/anomaly-det…

All Comments (21)
  • @colouredtime
    Thank you for sharing, very interesting lecture. Slides should be shown more often; it is difficult to follow the discussion without seeing the slides.
  • This is a nice intro to AD theory. Viewers need to see the slideshow more than the instructor himself though. Videographer please take note.
  • @shawn-jung
    Thank you for sharing the session video!
  • @anryxas1
    Very interesting area of ML. Thank you for the video
  • @MrChirros
    Very difficult to follow without slides. You need to show them more frequently and longer.
  • @loicverbeke216
    Is there a link where I can find the slides of the presentation ?
  • @usamarahman6178
    the lecture was really good but either show the slides, or atleast share the slides
  • @wei-chunlee7140
    These talks are great, but showing slides are usually more useful...
  • @alwaaffa
    You can help me with a master’s thesis for my software part (coding) in Python?
  • @sau002
    Where are the slides? They are very important. Becomes hard to follow the speaker.