Bundle Networks: Fiber Bundles, Local Trivializations, and a Generative Approach to Exploring Many-to-one Maps
Abstract
Many-to-one maps are ubiquitous in machine learning, from the image recognition model that assigns a multitude of distinct images to the concept of"cat"to the time series forecasting model which assigns a range of distinct time-series to a single scalar regression value. While the primary use of such models is naturally to associate correct output to each input, in many problems it is also useful to be able to explore, understand, and sample from a model's fibers, which are the set of input values such that , for fixed in the output space. In this paper we show that popular generative architectures are ill-suited to such tasks. Motivated by this we introduce a novel generative architecture, a Bundle Network, based on the concept of a fiber bundle from (differential) topology. BundleNets exploit the idea of a local trivialization wherein a space can be locally decomposed into a product space that cleanly encodes the many-to-one nature of the map. By enforcing this decomposition in BundleNets and by utilizing state-of-the-art invertible components, investigating a network's fibers becomes natural.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
1 top-tier paper has cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
Cited by top-tier papers1
Builds on1
Related papers
- CompoNet: Learning to Generate the Unseen by Part Synthesis and CompositionNadav Schor, Oren Katzir, Hao Zhang, Daniel Cohen-OrICCV 2019 · 63 citations
- Worldsheet: Wrapping the World in a 3D Sheet for View Synthesis from a Single ImageRonghang Hu, Nikhila Ravi, Alexander C. Berg, Deepak PathakICCV 2021 · 97 citations
- Bundle Neural Network for message diffusion on graphsJacob Bamberger, Federico Barbero, Xiaowen Dong, Michael M. BronsteinICLR 2025
- Lost in Latent Space: Examining failures of disentangled models at combinatorial generalisationMilton Llera Montero, Jeffrey S. Bowers, Rui Ponte Costa, Casimir J. H. Ludwig et al.NeurIPS 2022 · 29 citations
- Scalable Neural Network KernelsArijit Sehanobish, Krzysztof Marcin Choromanski, Yunfan Zhao, Kumar Avinava Dubey et al.ICLR 2024 · 9 citations
