Abstract
Black box characteristics of machine learning algorithms seriously hamper their application in the certain fields, such as medicine, military, finance and so on. So far, the interpretability of machine learning remains as a challenge. In this paper, we use Layer-wise Relevance Propagation (LRP) to calculate the relevance of the Convolutional Neural Network (CNN) on the input data, and visualize it as a heat map, so as to intuitively understand which features the Convolutional Neural Network are based on to make prediction, and then improve the model by analyzing the heat maps. In this article, by using the control variable method, the LRP algorithm is applied to the improved convolution neural network to obtain a new heat map. The difference between the two heat maps is analyzed to verify that the interpretable algorithm conforms to the objective facts. In this way, the interpretability of different network structures is compared and improved.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Silver, D., Huang, A., Maddison, C.J., Guez, A.: Mastering the game of Go with deep neural networks and tree search. Nature 529(7587), 484–489 (2016)
Chand, H.V., Karthikeyan, J.: CNN based driver drowsiness detection system using emotion analysis. Intell. Autom. Soft Comput. 31(2), 717–728 (2022)
Jin, L.L., Liang, H., Yang, C.S.: Sonar image recognition of underwater target based on convolutional neural network. J. Northwestern Polytech. Univ. 39(2), 285–291 (2021)
Cannizzaro, D., Aliberti, A., Bottaccioli, L., Macii, E.: solar radiation forecasting based on convolutional neural network and ensemble learning. Exp. Syst. Appl. 181, 115167 (2021)
Lee, C.C., Gao, Z.: Sign language recognition using two-stream convolutional neural networks with wi-fi signals. Appl. Sci. 10(24), 9005 (2020)
Hyun, J., Seong, H., Kim, E.: Universal pooling–a new pooling method for convolutional neural networks. Exp. Syst. Appl. 180, 115084 (2021)
Zhang, Z.Z., Zhou, W.X.: Image dehazing algorithm based on deep learning. J. South China Norm. Univ. (Nat. Sci. Edn.) 53(3), 123–128 (2019)
Miao, P., Srimahachota, T.: Cost-effective system for detection and quantification of concrete surface cracks by combination of convolutional neural network and image processing techniques. Constr. Build. Mater. 293, 123549 (2021)
Chen, P.: Research on the knowledge based of ship collision avoidance based on HSSVM and convolutional neural networks. Dalian Maritime University (2021)
Acevedo, A., Merino, A., Boldú, L., Molina, A.: A new convolutional neural network predictive model for the automatic recognition of hypogranulated neutrophils in myelodysplastic syndromes. Comput. Biol. Med. 134, 104479 (2021)
Khalili, E., Asl, B.M.: Automatic sleep stage classification using temporal convolutional neural network and new data augmentation technique from raw single-channel EEG. Comput. Methods Prog. Biomed. 204, 106063 (2021)
Yang, B., Cao, J.-M., Jiang, D.-P., Lv, J.-D.: Facial expression recognition based on dual-feature fusion and improved random forest classifier. Multim. Tools Appl. 77(16), 20477–20499 (2017)
Abu-Alhaija, M., Turab, N.M.: Automated learning of ecg streaming data through machine learning internet of things. Intell. Autom. Soft Comput. 32(1), 45–53 (2022)
Maheshwari, D., Ghosh, S.K., Tripathy, R.K., Sharma, M.: Automated accurate emotion recognition system using rhythm-specific deep convolutional neural network technique with multi-channel EEG signals. Comput. Biol. Med. 134, 104428 (2021)
Ding, G., et al.: Fish recognition using convolutional neural network. In: OCEANS 2017-Anchorage, pp. 1–4. IEEE (2017)
Shukla, A.K., Das, S.: Deep neural network and pseudo relevance feedback based query expansion. Comput. Mater. Continua 71(2), 3557–3570 (2022)
Zhu, Y.H., Jiang, Y.Z.: Optimization of face recognition algorithm based on deep learning multi feature fusion driven by big data. Image Vis. Comput. 104, 104023 (2020)
Böhle, M., Eitel, F., Weygandt, M., Ritter, K.: Layer-wise relevance propagation for explaining deep neural network decisions in MRI-based Alzheimer’s disease classification. Front. Aging Neurosci. 11, 194 (2020)
Leonid, T.T., Jayaparvathy, R.: Classification of elephant sounds using parallel convolutional neural network. Intell. Autom. Soft Comput. 32(3), 1415–1426 (2022)
Venkateswaran, N., Umadevi, K.: Hybridized wrapper filter using deep neural network for intrusion detection. Comput. Syst. Sci. Eng. 42(1), 1–14 (2022)
Jiang, Q.: It will take time for AI to win the highest level of human Go. Internet Weekly 4(6) (2016)
Zhen, H.: Artificial intelligence intervention in sentencing mechanism: dilemma, orientation and deconstruction. J. Chongqing Univ. (Soc. Sci. Edn.) (2020)
Zhao, P.: Application and development of artificial intelligence technology in clinical medical diagnosis. China New Telecommun. 21(22), 90–91 (2019)
Qiao, X., Xi, Y.: Artificial intelligence and the construction of modern economic system. Econ. Aspects 06, 81–91 (2018)
Zhu, M., Hou, J., Sun, S.: Domestic research progress of remote sensing image recognition based on deep learning. Surv. Geospat. Inf. 44(5), 67–73 (2021)
Samek, W., Binder, A., Montavon, G., Lapuschkin, S.: Evaluating the visualization of what a deep neural network has learned. IEEE Trans. Neural Netw. Learn. Syst. 28(11), 2660–2673 (2016)
Guan, C., Yang, Y.: Research on the application of back propagation neural network in social development. Comput. Times 5, 46–48 (2021)
Ke, Y., Lu, Y.: Pet recognition method based on vgg16. Electron. Prod. 21, 42–45 (2020)
Song, F.: Research on animal facial recognition algorithm based on deep learning. Hangzhou Dianzi University (2019)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Jia, W., Zhang, S., Jiang, Y., Xu, L. (2022). Interpreting Convolutional Neural Networks via Layer-Wise Relevance Propagation. In: Sun, X., Zhang, X., Xia, Z., Bertino, E. (eds) Artificial Intelligence and Security. ICAIS 2022. Lecture Notes in Computer Science, vol 13338. Springer, Cham. https://doi.org/10.1007/978-3-031-06794-5_37
Download citation
DOI: https://doi.org/10.1007/978-3-031-06794-5_37
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-06793-8
Online ISBN: 978-3-031-06794-5
eBook Packages: Computer ScienceComputer Science (R0)