arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1912.00155cs.LGstat.ML
收藏

解纠缠挑战:从正规化到重构

Disentanglement Challenge: From Regularization to Reconstruction

Guangdong University of Technology(广东工业大学);Carnegie Mellon University(卡内基梅隆大学)

Jie Qiao, Zijian Li, Boyan Xu, Ruichu Cai, Kun Zhang

摘要:

学习解纠缠表示的挑战最近吸引了很多关注,并且归结为使用新的真实世界解纠缠数据集的竞争(Gondal等人,2019年)。已经提出了各种基于变分自动编码器的方法来解决这个问题,通过加强表示之间的独立性和修改变分下界中的正则化项。然而,Locatello等人(2018)最近的工作表明,所提出的方法受到随机性和超参数选择的严重影响。在这项工作中,而不是设计一个新的正则化项,我们采用FactorVAE,但改善了重建性能,增加了网络的容量和训练步骤。该策略被证明是非常有效的,并在挑战中获得第一名。

英文摘要:

The challenge of learning disentangled representation has recently attracted much attention and boils down to a competition using a new real world disentanglement dataset (Gondal et al., 2019). Various methods based on variational auto-encoder have been proposed to solve this problem, by enforcing the independence between the representation and modifying the regularization term in the variational lower bound. However recent work by Locatello et al. (2018) has demonstrated that the proposed methods are heavily influenced by randomness and the choice of the hyper-parameter. In this work, instead of designing a new regularization term, we adopt the FactorVAE but improve the reconstruction performance and increase the capacity of network and the training step. The strategy turns out to be very effective and achieve the 1st place in the challenge.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑