[1912.05827] An Efficient Explorative Sampling Considering the Generative Boundaries of Deep Generative Neural Networkscontact arXivarXiv Twitter

Deep generative neural networks (DGNNs) have achieved realistic and high-quality data generation. In particular, the adversarial training scheme has been applied to many DGNNs and has exhibited powerful performance. Despite of recent advances in generative networks, identifying the image generation mechanism still remains challenging. In this paper, we present an explorative sampling algorithm to analyze generation mechanism of DGNNs. Our method efficiently obtains samples with identical attributes from a query image in a perspective of the trained model. We define generative boundaries which determine the activation of nodes in the internal layer and probe inside the model with this information. To handle a large number of boundaries, we obtain the essential set of boundaries using optimization. By gathering samples within the region surrounded by generative boundaries, we can empirically reveal the characteristics of the internal layers of DGNNs. We also demonstrate that our algorith

1 mentions: @kchonyc
Date: 2020/02/13 20:21

Referring Tweets

@kchonyc very cool work on understanding deep generators by Giyoung Jeon, Haedong Jeong and Jaesik Choi: t.co/i3SpaImo0p t.co/VAdup1Hnwz

Related Entries

Read more GitHub - victordibia/handtracking: Building a Real-time Hand-Detector using Neural Networks (SSD) on...
0 users, 1 mentions 2019/11/09 15:51
Read more Neural networks for Graph Data NeurIPS2018読み会@PFN
25 users, 9 mentions 2019/01/26 09:46
Read more [1911.12116] Analysis of Explainers of Black Box Deep Neural Networks for Computer Vision: A Survey
0 users, 4 mentions 2019/11/28 23:20
Read more GitHub - thunlp/GNNPapers: Must-read papers on graph neural networks (GNN)
1 users, 0 mentions 2019/08/18 08:16
Read more Deep Forest :Deep Neural Networkの代替へ向けて - QiitaQiita
0 users, 0 mentions 2018/04/25 17:22