{"type":"ScholarlyArticle","title":"Unifying and Merging Well-trained Deep Neural Networks for Inference Stage","authors":[{"name":"Yi-Min Chou","slug":"yi-min-chou","given":"Yi-Min","family":"Chou","affiliation":"Institute of Information Science , Academia Sinica , Taipei , Taiwan; MOST Joint Research Center for AI Technology and All Vista Healthcare"},{"name":"Yi-Ming Chan","slug":"yi-ming-chan","given":"Yi-Ming","family":"Chan","affiliation":"Institute of Information Science , Academia Sinica , Taipei , Taiwan; MOST Joint Research Center for AI Technology and All Vista Healthcare"},{"name":"Jia-Hong Lee","slug":"jia-hong-lee","given":"Jia-Hong","family":"Lee","affiliation":"Institute of Information Science , Academia Sinica , Taipei , Taiwan; MOST Joint Research Center for AI Technology and All Vista Healthcare"},{"name":"Chih-Yi Chiu","slug":"chih-yi-chiu","given":"Chih-Yi","family":"Chiu","affiliation":"Department of Computer Science and Information Engineering , National Chiayi University , Taiwan; Conv Conv Conv Conv Conv Conv Conv FC FC"},{"name":"Chu-Song Chen","slug":"chu-song-chen","given":"Chu-Song","family":"Chen","affiliation":"Institute of Information Science , Academia Sinica , Taipei , Taiwan; MOST Joint Research Center for AI Technology and All Vista Healthcare"}],"abstract":"We propose a novel method to merge convolutional neural-nets for the inference stage. Given two well-trained networks that may have different architectures that handle different tasks, our method aligns the layers of the original networks and merges them into a unified model by sharing the representative codes of weights.\r\nThe shared weights are further re-trained to fine-tune the performance of the merged model. The proposed method effectively produces a compact model that may run original tasks simultaneously on resource-limited devices. As it preserves the general architectures and leverages the co-used weights of well-trained networks, a substantial training overhead can be reduced to shorten the system development time. Experimental results demonstrate a satisfactory performance and validate the effectiveness of the method.","doi":"10.24963/ijcai.2018/283","url":"https://ijcai.org/proceedings/2018/283","pdf":"https://ijcai.org/proceedings/2018/0283.pdf","markdown":"https://ijcai.org/proceedings/2018/283.md","bibtex":"https://ijcai.org/proceedings/2018/bibtex/283","volume":{"slug":"2018","title":"Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2018","url":"https://ijcai.org/proceedings/2018/","year":2018,"location":"Stockholm","dates":"13-19 July 2018","editors":"Jérôme Lang","isbn":"978-0-9992411-2-7","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2018"},"section":"Machine Learning","track":"Main track","number":283,"pages":{"first":2049,"last":2056},"keywords":["Computer Vision: Computer Vision","Machine Learning: Neural Networks","Machine Learning: Deep Learning"],"videos":[],"references":[{"position":1,"raw":"[Abadi et al., 2016] Martín Abadi, Ashish Agarwal, Paul Barham, Eugene Brevdo, Zhifeng Chen, Craig Citro, Greg S Corrado, Andy Davis, Jeffrey Dean, Matthieu Devin, et al. Tensorflow: Large-scale machine learning on heterogeneous distributed systems. arXiv, 2016.","title":"Tensorflow: Large-scale machine learning on heterogeneous distributed systems","authors":"Martín Abadi, Ashish Agarwal, Paul Barham, Eugene Brevdo, Zhifeng Chen, Craig Citro, Greg S Corrado, Andy Davis, Jeffrey Dean, Matthieu Devin","year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Anwar et al., 2017] Sajid Anwar, Kyuyeon Hwang, and Wonyong Sung. Structured pruning of deep convolutional neural networks. ACM Journal on Emerging Technologies in Computing Systems (JETC), 13(3):32, 2017.","title":"Structured pruning of deep convolutional neural networks","authors":"Sajid Anwar, Kyuyeon Hwang, Wonyong Sung","year":"2017","venue":"ACM Journal on Emerging Technologies in Computing Systems (JETC)","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Keogh, 2012] Yuan Hao Bilson Campana Eamonn Keogh. Monitoring and mining insect sounds in visual space. 2012.","title":"Monitoring and mining insect sounds in visual space","authors":"Yuan Hao, Bilson Campana, Eamonn Keogh","year":"2012","venue":null,"doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[LeCun et al., 1998] Yann LeCun, Léon Bottou, Yoshua Bengio, and Patrick Haffner. Gradient-based learning ap- plied to document recognition. Proceedings of the IEEE, 86(11):2278-2324, 1998.","title":"Gradient-based learning applied to document recognition","authors":"Yann Lecun, Léon Bottou, Yoshua Bengio, Patrick Haffner","year":"1998","venue":"Proceedings of the IEEE","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Li et al., 2016] Hao Li, Asim Kadav, Igor Durdanovic, Hanan Samet, and Hans Peter Graf. Pruning filters for ef- ficient convnets. arXiv preprint arXiv:1608.08710, 2016.","title":"Pruning filters for efficient convnets","authors":"Hao Li, Asim Kadav, Igor Durdanovic, Hanan Samet, Hans Peter Graf","year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Liu et al., 2016] Kuan-Hsien Liu, Ting-Yen Chen, and Chu- Song Chen. Mvc: A dataset for view-invariant clothing re- trieval and attribute prediction. In Proceedings of ACM on International Conference on Multimedia Retrieval, pages 313-316, 2016.","title":"Mvc: A dataset for view-invariant clothing retrieval and attribute prediction","authors":"Kuan-Hsien, Ting-Yen Liu, Chu-Song Chen, Chen","year":"2016","venue":"Proceedings of ACM on International Conference on Multimedia Retrieval","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Molchanov et al., 2016] Pavlo Molchanov, Stephen Tyree, Tero Karras, Timo Aila, and Jan Kautz. Pruning convo- lutional neural networks for resource efficient inference. 2016.","title":"Pruning convolutional neural networks for resource efficient inference","authors":"Pavlo Molchanov, Stephen Tyree, Tero Karras, Timo Aila, Jan Kautz","year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Rastegari et al., 2016] Mohammad Rastegari, Vicente Or- donez, Joseph Redmon, and Ali Farhadi. Xnor-net: Ima- genet classification using binary convolutional neural net- works. In European Conference on Computer Vision, pages 525-542. Springer, 2016.","title":"Xnor-net: Imagenet classification using binary convolutional neural networks","authors":"Mohammad Rastegari, Vicente Ordonez, Joseph Redmon, Ali Farhadi","year":"2016","venue":"European Conference on Computer Vision","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Ruder, 2017] Sebastian Ruder. An overview of multi- task learning in deep neural networks. arXiv preprint arXiv:1706.05098, 2017.","title":"An overview of multitask learning in deep neural networks","authors":"Sebastian Ruder","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Simonyan and Zisserman, 2014] Karen Simonyan and An- drew Zisserman. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556, 2014.","title":"Very deep convolutional networks for large-scale image recognition","authors":"Karen Simonyan, Andrew Zisserman","year":"2014","venue":null,"doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Wang et al., 2013] Qian Wang, Xianyi Zhang, Yunquan Zhang, and Qing Yi. Augem: automatically generate high performance dense linear algebra kernels on x86 cpus. In Proceedings of the International Conference on High Per- formance Computing, Networking, Storage and Analysis, page 25. ACM, 2013.","title":"Augem: automatically generate high performance dense linear algebra kernels on x86 cpus","authors":"Qian Wang, Xianyi Zhang, Yunquan Zhang, Qing Yi","year":"2013","venue":"Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Wu et al., 2016] Jiaxiang Wu, Cong Leng, Yuhang Wang, Qinghao Hu, and Jian Cheng. Quantized convolutional neural networks for mobile devices. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recog- nition, pages 4820-4828, 2016.","title":"Quantized convolutional neural networks for mobile devices","authors":"Jiaxiang Wu, Cong Leng, Yuhang Wang, Qinghao Hu, Jian Cheng","year":"2016","venue":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Xiao et al., 2017] Han Xiao, Kashif Rasul, and Roland Vollgraf. Fashion-mnist: a novel image dataset for bench- marking machine learning algorithms, 2017.","title":"Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms","authors":"Han Xiao, Kashif Rasul, Roland Vollgraf","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Yang et al., 2018] Huei-Fang Yang, Kevin Lin, and Chu- Song Chen. Supervised learning of semantics-preserving hash via deep convolutional neural networks. IEEE trans- actions on pattern analysis and machine intelligence, 40(2):437-451, 2018.","title":"Supervised learning of semantics-preserving hash via deep convolutional neural networks","authors":"Huei-Fang Yang, Kevin Lin, Chu-Song Chen","year":"2018","venue":"IEEE transactions on pattern analysis and machine intelligence","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Zeiler and Fergus, 2014] Matthew D Zeiler and Rob Fergus. Visualizing and understanding convolutional networks. In Proceedings of the European Conference on Computer Vi- sion, pages 818-833, 2014.","title":"Visualizing and understanding convolutional networks","authors":"D Matthew, Rob Zeiler, Fergus","year":"2014","venue":"Proceedings of the European Conference on Computer Vision","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Zhang et al., 2012] Xianyi Zhang, Qian Wang, and Yun- quan Zhang. Model-driven level 3 blas performance opti- mization on loongson 3a processor. In IEEE International Conference on Parallel and Distributed Systems, pages 684-691. IEEE, 2012.","title":"Model-driven level 3 blas performance optimization on loongson 3a processor","authors":"Xianyi Zhang, Qian Wang, Yunquan Zhang","year":"2012","venue":"IEEE International Conference on Parallel and Distributed Systems","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}