paper-with-me

Papers

Topological Obstructions and How to Avoid Them

2023-12-12 · NeurIPS 2023 11 · Babak Esmaeili, Robin Walters, Heiko Zimmermann, Jan-Willem van de Meent

Incorporating geometric inductive biases into models can aid interpretability and generalization, but encoding to a specific geometric structure can be challenging due to the imposed topological constraints. In this paper, we theoretically and empirically characterize obstructions to training encoders with geometric latent spaces. We show that local optima can arise due to singularities (e.g. self-intersection) or due to an incorrect degree or winding number. We then discuss how normalizing flows can potentially circumvent these obstructions by defining multimodal variational distributions. Inspired by this observation, we propose a new flow-based model that maps data points to multimodal distributions over geometric spaces and empirically evaluate our model on 2 domains. We observe improved stability during training and a higher chance of converging to a homeomorphic encoder.

📄 PDF Abstract BibTeX arXiv:2312.07529

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Normalizing Flows Normalizing Flows are a method for constructing complex distributions by transforming a probability density through a series of invertible mappings. By repeatedly applying…

Similar Papers 제목 키워드 기반

Topological obstructions in neural networks learning

2020-12-31 · Serguei Barannikov, Daria Voronkova, Ilya Trofimov, Alexander Korotin 외

We apply topological data analysis methods to loss functions to gain insights into learning of deep neural networks and deep neural networks generalization properties. We use the Morse complex of the loss function to rel…

Topological Data Analysis

A Functorial Formulation of Neighborhood Aggregating Deep Learning

2026-04-27 · Sun Woo Park, Yun Young Choi, U Jin Choi, Youngho Woo arxiv

We provide a mathematical interpretation of convolutional (or message passing) neural networks by using presheaves and copresheaves of the set of continuous functions over a topological space. Based on this interpretatio…

Cohomological Obstructions to Global Counterfactuals: A Sheaf-Theoretic Foundation for Generative Causal Models

2026-03-18 · Rui Wu, Hong Xie, Yongjun Li arxiv

Current continuous generative models (e.g., Diffusion Models, Flow Matching) implicitly assume that locally consistent causal mechanisms naturally yield globally coherent counterfactuals. In this paper, we prove that thi…

Universal Joint Approximation of Manifolds and Densities by Simple Injective Flows

2021-10-08 · Michael Puthawala, Matti Lassas, Ivan Dokmanić, Maarten de Hoop

We study approximation of probability measures supported on $n$-dimensional manifolds embedded in $\mathbb{R}^m$ by injective flows -- neural networks composed of invertible flows and injective layers. We show that in ge…

Diffusion Variational Autoencoders

2019-01-25 · Luis A. Pérez Rey, Vlado Menkovski, Jacobus W. Portegies

A standard Variational Autoencoder, with a Euclidean latent space, is structurally incapable of capturing topological properties of certain datasets. To remove topological obstructions, we introduce Diffusion Variational…