BriefGPT.xyz
Feb, 2022
无监督学习中可识别性的陷阱：关于《表示学习的期望：因果透视》的注解
On Pitfalls of Identifiability in Unsupervised Learning. A Note on: "Desiderata for Representation Learning: A Causal Perspective"
HTML
PDF
Shubhangi Ghosh, Luigi Gresele, Julius von Kügelgen, Michel Besserve, Bernhard Schölkopf
TL;DR
本文讨论了一种可能的未标识结果，说明了非线性独立成分分析理论基础上的建构，并通过适当构造的反例说明了表示学习中的其他反例及其可识别性。
Abstract
model identifiability
is a desirable property in the context of
unsupervised representation learning
. In absence thereof, different models may be observationally indistinguishable while yielding representations t
→