neural network approach to discovering temporal correlations
DESCRIPTION
Neural Network Approach to Discovering Temporal Correlations. S.A.Dolenko, Yu.V.Orlov, I.G.Persiantsev, Ju.S.Shugai Scobeltsyn Institute of Nuclear Physics, Moscow State University E-mail: [email protected]. Statement of the problem. - PowerPoint PPT PresentationTRANSCRIPT
Neural Network Approach to Discovering Temporal
Correlations
S.A.Dolenko, Yu.V.Orlov, I.G.Persiantsev, Ju.S.Shugai
Scobeltsyn Institute of Nuclear Physics,Moscow State University
E-mail: [email protected]
Statement of the problem• Discovering causal relationship “behavior - event”
- What type of behavior has initiated the event?- What phenomenon has initiated the event?
• Application - geomagnetic storms forecasting;SOHO - http://sohowww.nasacom.nasa.gov
• Complexity of the task- What is the delay between the event and the
moment of its initiation?- Can use “passive observation” only
Objective of the research:
Development of an algorithm for discovering temporary correlations
Model assumptions• Data = Sequence of scene images• Scene = Set of objects• Lifetime of objects >> Registration rate• Object = Set of features• Phenomenon = Unknown combination of features• Event:
- Initiated by unknown phenomenon within “Initiation duration”
- Search interval >> Initiation duration - Limited number of events’ types- Fixed (unknown) delay for a given type of event
Find the most probable phenomenon and delay
Scheme of the algorithm
Model experiment 1: Single event
Model experiment 2: Two events
Approaching the Sun...
Future development
• NN experts specialization through competition• Second hierarchical level - NN Supervisor
• Discovering temporal correlations “Sun surface - Geomagnetic storms”- Increasing forecast horizon- Improving forecast reliability
• Applications in seismology, medicine, finance,…