paper-with-me

홈 › Papers

Analyzing the Latent Space of GAN through Local Dimension Estimation

2022-05-26 · Jaewoong Choi, Geonho Hwang, Hyunsoo Cho, Myungjoo Kang

The impressive success of style-based GANs (StyleGANs) in high-fidelity image synthesis has motivated research to understand the semantic properties of their latent spaces. In this paper, we approach this problem through a geometric analysis of latent spaces as a manifold. In particular, we propose a local dimension estimation algorithm for arbitrary intermediate layers in a pre-trained GAN model. The estimated local dimension is interpreted as the number of possible semantic variations from this latent variable. Moreover, this intrinsic dimension estimation enables unsupervised evaluation of disentanglement for a latent space. Our proposed metric, called Distortion, measures an inconsistency of intrinsic tangent space on the learned latent space. Distortion is purely geometric and does not require any additional attribute information. Nevertheless, Distortion shows a high correlation with the global-basis-compatibility and supervised disentanglement score. Our work is the first step towards selecting the most disentangled latent space among various latent spaces in a GAN without attribute labels.

📄 PDF Abstract BibTeX arXiv:2205.13182

Code (0)

등록된 구현이 없습니다.

Tasks

AttributeDisentanglementImage Generation

Methods 이 논문이 사용한 방법론

HuMan(Expedia)||How do I get a human at Expedia? How do I get a human at Expedia? How Do I Get a Human at Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Real-Time Help & Exclusive…
StyleGAN 설명 없음
Convolution A convolution is a type of matrix operation, consisting of a kernel, a small matrix of weights, that slides over input data performing element-wise multiplication with the…
R1 Regularization R_INLINE_MATH_1 Regularization is a regularization technique and gradient penalty for training [generative adversarial…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
PCA Principle Components Analysis (PCA) is an unsupervised method primary used for dimensionality reduction within machine learning. PCA is calculated via a singular value…
Adaptive Instance Normalization 설명 없음
Feedforward Network A Feedforward Network, or a Multilayer Perceptron (MLP), is a neural network with solely densely connected layers. This is the classic neural network architecture of the…

Similar Papers 제목 키워드 기반

Prediction, Consistency, Curvature: Representation Learning for Locally-Linear Control

2019-09-04 · ICLR 2020 1 · Nir Levine, Yin-Lam Chow, Rui Shu, Ang Li 외

Many real-world sequential decision-making problems can be formulated as optimal control with high-dimensional observations and unknown dynamics. A promising approach is to embed the high-dimensional observations into a …

Decision MakingOpen-Ended Question AnsweringRepresentation LearningSequential Decision Making

Knowledge-Embedded Latent Projection for Robust Representation Learning

2026-02-18 · Weijing Tang, Ming Yuan, Zongqi Xia, Tianxi Cai arxiv

Latent space models are widely used for analyzing high-dimensional discrete data matrices, such as patient-feature matrices in electronic health records (EHRs), by capturing complex dependence structures through low-dime…

Representation Learning

HALO: Hybrid Auto-encoded Locomotion with Learned Latent Dynamics, Poincaré Maps, and Regions of Attraction

2026-04-20 · Blake Werner, Sergio A. Esteban, Massimiliano De Sa, Max H. Cohen 외 arxiv

Reduced-order models are powerful for analyzing and controlling high-dimensional dynamical systems. Yet constructing these models for complex hybrid systems such as legged robots remains challenging. Classical approaches…

Discovering Stylistic Variations in Distributional Vector Space Models via Lexical Paraphrases

2017-09-01 · WS 2017 9 · Xing Niu, Marine Carpuat

Detecting and analyzing stylistic variation in language is relevant to diverse Natural Language Processing applications. In this work, we investigate whether salient dimensions of style variations are embedded in standar…

Semantic Textual SimilarityWord Embeddings

Thinner Latent Spaces: Detecting dimension and imposing invariance through autoencoder gradient constraints

2024-08-28 · George A. Kevrekidis, Mauro Maggioni, Soledad Villar, Yannis G. Kevrekidis

Conformal Autoencoders are a neural network architecture that imposes orthogonality conditions between the gradients of latent variables towards achieving disentangled representations of data. In this letter we show that…