Machine learning explanation can significantly boost machine learning's application in decision making,but the usability of current methods is limited in human-centric explanation,especially for transfer learning,an important machine learning branch that aims at utilizing knowledge from one learning domain (i.e., a pair of dataset and prediction task) to enhance prediction model training in another learning domain.In this paper, we propose an ontology-based approach for human-centric explanation of transfer learning. Three kinds of knowledge-based explanatory evidence, with different granularities, including general factors, particular narrators and core contexts are first proposedand then inferred with both local ontologies and external knowledge bases.The evaluation with US flight data and DBpedia has presented their confidence and availability in explaining the transferability of feature representation in flight departure delay forecasting.
|Title of host publication||Principles of Knowledge Representation and Reasoning|
|Subtitle of host publication||Proceedings of the Sixteenth International Conference (KR2018)|
|Editors||Michael Thielscher, Francesca Toni, Frank Wolter|
|Place of Publication||Palo Alto, California|
|Number of pages||10|
|Publication status||Published - 24 Sept 2018|
|Event||16th International Conference on Principles of Knowledge Representation and Reasoning - Tempe, United States|
Duration: 30 Oct 2018 → 2 Nov 2018
|Conference||16th International Conference on Principles of Knowledge Representation and Reasoning|
|Period||30/10/18 → 2/11/18|
The work was partially funded by the research center SIRIUS, the EPSRC project DBOnto, the EU Marie Curie KDrive project (286348) and NSFC61673338.
- transfer learning
- description logic
- explanative AI