Explaining neural network reasoning
Aus International Center for Computational Logic
Explaining neural network reasoning
Verteidigung Studienarbeit von Walter Forkel
- Veranstaltungsort: APB 3027
- Beginn: 3. März 2016 um 13:30
- Ende: 3. März 2016 um 14:30
- Forschungsgruppe: Computational Logic
- Event series: KBS Seminar
- iCal
Neural networks are in heavy use everywhere in artificial intelligence today. They produce state of the art results in many different domains. Still the lack of comprehensibility of neural networks makes them difficult to use in certain domains where human understandable explanations are needed. In this talk we present different approaches on how to automatically generate explanations of the neural networks reasoning.
