From 0 to Anomaly Detection in your infrastructure metrics in 15 minutes

Using the NuPIC framework (https://github.com/numenta/nupic/tree...), we will show the basics of Anomaly Detection using HTM ( Hierarchical Temporal Memory), and perform a demo trying to detect an anomaly in an infrastructure metric (such as a host CPU).

Transcript

1. Alejandro Guirao
   @lekum
   From 0 to anomaly detection in your
   infrastructure metrics in 15 minutes
   github.com/lekum
   

   View Slide

2. “
   The checks are often inflexible Boolean logic or arbitrary static in time thresholds. They generally
   rely on a specific result or range being matched.
   The checks again don’t consider the dynamism of most complex systems.
   A match or a breach in a threshold may be important or could have been triggered by an
   exceptional event—or it could even be a natural consequence of growth.
   James Turnbull (“The Art of Monitoring”)
   

   View Slide

3. ◉ Biologically constrained theory of machine intelligence
   ◉ “On Intelligence” (2004, Jeff Hawkins)
   ◉ Numenta (2005)
   Time-based learning algorithms that store and
   recall temporal patterns
   Hierarchical Temporal Model (HTM)
   

   View Slide

4. ◉ Encoders
   ◉ Spatial Poolers -> Sparsely Dense Representation
   ◉ Temporal Pooler -> Cortical Learning Algorithm
   HTM concepts
   

   View Slide

5. Cortical Learning Algorithm
   

   View Slide

6. ◉ NuPIC
   ◉ C++/Python/Java/Clojure
   ◉ HTM Studio for Anomaly Detection
   ◉ Grok
   HTM implementations
   

   View Slide

7. ◉ https://github.com/numenta/nupic/wiki/Learning-NuPIC
   ◉ http://numenta.org/htm-school/ (Videos)
   ◉ https://github.com/lekum/htm-demo
   Further reading
   

   View Slide

8. Happy hacking!
   Alejandro Guirao
   @lekum
   lekum.org
   github.com/lekum
   

   View Slide