paper-with-me

홈 › Papers

Benchmarks, Algorithms, and Metrics for Hierarchical Disentanglement

2021-02-09 · Andrew Slavin Ross, Finale Doshi-Velez

In representation learning, there has been recent interest in developing algorithms to disentangle the ground-truth generative factors behind a dataset, and metrics to quantify how fully this occurs. However, these algorithms and metrics often assume that both representations and ground-truth factors are flat, continuous, and factorized, whereas many real-world generative processes involve rich hierarchical structure, mixtures of discrete and continuous variables with dependence between them, and even varying intrinsic dimensionality. In this work, we develop benchmarks, algorithms, and metrics for learning such hierarchical representations.

📄 PDF Abstract BibTeX arXiv:2102.05185

Code (1)

dtak/hierarchical-disentanglement 공식 구현 tf

Tasks

DisentanglementRepresentation Learning

Similar Papers 제목 키워드 기반

Progressive Learning and Disentanglement of Hierarchical Representations

2020-02-24 · ICLR 2020 1 · Zhiyuan Li, Jaideep Vitthal Murkute, Prashnna Kumar Gyawali, Linwei Wang

Learning rich representation from data is an important task for deep generative models such as variational auto-encoder (VAE). However, by extracting high-level abstractions in the bottom-up inference process, the goal o…

DisentanglementRepresentation Learning

Variational Learning with Disentanglement-PyTorch

2019-12-11 · Amir H. Abdi, Purang Abolmaesumi, Sidney Fels

Unsupervised learning of disentangled representations is an open problem in machine learning. The Disentanglement-PyTorch library is developed to facilitate research, implementation, and testing of new variational algori…

DisentanglementScheduling

Variational Learning with Disentanglement-PyTorch

2019-11-15 · NeurIPS Workshop DC_S2 2019 12 · Anonymous

Unsupervised learning of disentangled representations is an open problem in machine learning. The Disentanglement-PyTorch library is developed to facilitate research, implementation, and testing of new variational algori…

DisentanglementScheduling

How to Not Measure Disentanglement

2019-10-12 · Anna Sepliarskaia, Julia Kiseleva, Maarten de Rijke

To evaluate disentangled representations several metrics have been proposed. However, theoretical guarantees for conventional metrics of disentanglement are missing. Moreover, conventional metrics do not have a consisten…

DisentanglementReinforcement LearningTransfer LearningZero-Shot Learning

Boxhead: A Dataset for Learning Hierarchical Representations

2021-10-07 · NeurIPS Workshop SVRHM 2021 12 · Yukun Chen, Andrea Dittadi, Frederik Träuble, Stefan Bauer 외

Disentanglement is hypothesized to be beneficial towards a number of downstream tasks. However, a common assumption in learning disentangled representations is that the data generative factors are statistically independe…

Disentanglement