Abstract | ||
---|---|---|
We address the interpretability of convolutional neural networks (CNNs) for predicting a geo-location from an image. In a pilot experiment we classify images of Pittsburgh vs Tokyo and visualize the learned CNN filters. We found that varying the CNN architecture leads to variating in the visualized filters. This calls for further investigation of the effective parameters on the interpretability of CNNs. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1109/eScience.2018.00125 | 2018 IEEE 14TH INTERNATIONAL CONFERENCE ON E-SCIENCE (E-SCIENCE 2018) |
Keywords | Field | DocType |
convolutional neural network (CNN), interpretability, place recognition, visualization, classification | Data mining,Interpretability,Architecture,Intelligent decision support system,Convolutional neural network,Computer science,Visualization,Sight,Artificial intelligence,Artificial neural network,Deep neural networks | Conference |
ISSN | Citations | PageRank |
2325-372X | 0 | 0.34 |
References | Authors | |
0 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Seyran Khademi | 1 | 3 | 2.26 |
Xiangwei Shi | 2 | 0 | 1.35 |
Tino Mager | 3 | 0 | 0.34 |
Ronald Siebes | 4 | 2 | 1.41 |
Carola Hein | 5 | 0 | 0.34 |
Victor de Boer | 6 | 181 | 29.78 |
Jan C. van Gemert | 7 | 1505 | 98.97 |