This work proposes a model for geodesic distances and flows on manifolds.
problem Geodesic distances and flows on differentiable manifolds.
method Manifold-augmented Eikonal equation solutions.
result Geodesic flow provides globally length-minimizing curves.
Neural Manifold ODEs improve manifold data modeling.
problem Adapting deep generative models to non-Euclidean spaces.
method Introducing Neural Manifold ODEs for manifold generalization and continuous probability computation.
result Improves density estimation and downstream tasks on arbitrary manifolds.
Deep Gaussian processes on manifolds improve performance on complex data.
problem Complex data on manifolds that shallow models struggle with.
method Residual deep Gaussian processes on Riemannian manifolds.
result Significant improvement in prediction quality and uncertainty calibration.
Unique minimal model for LCK manifolds proved.
problem Characterizing unique minimal models for LCK manifolds.
method Proving bimeromorphic maps are holomorphic.
result LCK manifolds have a unique minimal model.
M-flows learn data manifolds and densities, improving manifold learning and inference.
problem Representing datasets with manifold structure more faithfully.
method Combining normalizing flows, GANs, autoencoders, and energy-based models, with a new training algorithm.
result M-flows learn data manifolds better than standard flows and provide handles for dimensionality reduction.
A new logit model derived from the Weibull manifold.
problem No potential function on the Weibull manifold.
method Extracted a logit model from the two-parameter Weibull model.
result Found a completely integrable Hamiltonian gradient system on the logit model.
Generative mixture models of VAEs learn manifolds for inverse problems.
problem Representing high-dimensional data manifolds efficiently and accurately.
method Mixture model of variational autoencoders (VAEs) with Riemannian gradient descent.
result Learned manifold enables solving inverse problems with data fidelity.
Proposes a new flow model to better represent data on manifolds.
problem Flow models struggle to represent data on lower-dimensional manifolds accurately.
method Introduces a manifold prior that leverages spread divergence to improve model performance.
result Improves both sample and representation quality, identifies manifold intrinsic dimension.
This work proposes a new neural implicit manifold model for more accurate density estimation on manifolds.
problem Current generative models struggle with representing manifolds accurately and learning densities within them.
method Proposes a neural implicit manifold model and a constrained energy-based model to learn manifold-supported distributions.
result The proposed model can learn manifold-supported distributions with complex topologies more accurately than pushforward models.
Equivariant flows learn symmetrical distributions on manifolds.
problem Learning symmetrical distributions on arbitrary manifolds.
method Equivariant manifold flows.
result Learned gauge invariant densities over SU(n) in quantum field theory.
ManifoldShap improves model explanations by restricting evaluations to the data manifold.
problem Inaccurate and misleading model explanations due to reliance on out-of-distribution data.
method Restricts model evaluations to the data manifold to avoid off-manifold perturbations.
result ManifoldShap provides more accurate and intuitive explanations than existing methods.
The paper tackles manifold overfitting in deep generative models.
problem Manifold overfitting occurs when generative models learn the manifold itself instead of the distribution on it.
method The authors propose a two-step procedure: dimensionality reduction followed by maximum-likelihood density estimation.
result The two-step procedure avoids manifold overfitting and enables density estimation on learned manifolds.
A diffusion model estimates data manifold dimension by tracking likelihood increases.
problem Estimating the intrinsic dimension of data manifolds.
method Trained diffusion model approximates score function, revealing manifold directionality.
result Diffusion model provides an approximation of the tangent space's dimension.
Generative model on manifolds reduces divergence computation and improves scalability.
problem Difficulties in modeling data on non-Euclidean spaces due to expensive divergence computation and approximations of heat kernel.
method Riemannian Diffusion Mixture, a principled framework using a mixture of bridge processes.
result Achieves superior performance on diverse manifolds with reduced simulation steps.
Paper introduces Manifold Probe for discovering representation manifolds in superposition.
problem Discovering representation manifolds in complex superposition representations.
method Generalizes linear regression probes to learn feature spaces and directions in superposition representations.
result Demonstrates Manifold Probe on Llama 2-7b representations, finding causally involved manifolds in model behaviour.
RSGMs extend SGMs to Riemannian manifolds for better data modeling.
problem Current SGMs are limited to Euclidean spaces; RSGMs handle Riemannian manifolds.
method RSGMs use a noising stage with a diffusion process and a denoising model approximating the time-reversal of the diffusion on Riemannian manifolds.
result RSGMs improve generative modeling for data on Riemannian manifolds.
Minimal dimensions found for flag manifolds embeddings.
problem Finding the smallest dimensions for flag manifolds embeddings.
method Equivariant embeddings of orthogonal and unitary groups acting on real and complex flag manifolds.
result Minimal dimensions achieved at isospectral models.
VAELLS learns latent manifold structure to improve VAE model accuracy.
problem VAEs struggle with mismatched latent structure and global structure.
method Integrates learnable manifold model into latent space of VAE.
result Improves model accuracy by matching prior to data manifold structure.
Study spin chains and sigma models on flag manifolds, calculating spectra and geodesics.
problem Understanding the spectrum and geodesics of sigma models on flag manifolds.
method Connecting SU(n) spin chains to sigma models and calculating spectra and geodesics.
result Calculated the spectrum of the Laplace-Beltrami operator and geodesics for CP1 and F3. Optimizes functions on manifolds using Gaussian processes and graph models.
problem Optimizing functions on unknown manifolds with limited data.
method Graph Gaussian process surrogate model for sequential optimization.
result Established regret bounds for the proposed algorithm.
We introduce the Locally Linear Latent Variable Model (LL-LVM), a probabilistic model for non-linear manifold discovery that describes a joint distribution over observations, their manifold coordinates and locally linear maps conditioned on a set of neighbourhood relationships. The model allows straightforward variatio…
Develops a method for manifold learning with small sample size datasets.
problem Improving manifold learning performance for multiple tasks with limited samples.
method Uses instance and model transfer to integrate manifold models from similar tasks.
result Successfully estimates manifolds with tiny sample sizes across multiple tasks.
Improved diffusion map enhances manifold regularization for semi-supervised learning.
problem Limited performance of manifold regularization models in capturing global structure.
method Enhanced diffusion map with improved label propagation function.
result Proposed method improves manifold regularization model's performance.
Efficient diffusion model for symmetric manifolds reduces training and computation costs.
problem Heat kernel computations for manifold diffusion models are computationally expensive and infeasible.
method Spatially-varying covariance diffusion model, efficient objective derived via Ito's Lemma.
result Our model reduces training time and arithmetic operations by orders of magnitude.
We show any Riemannian curvature model can be geometrically realized by a manifold with constant scalar curvature. We also show that any pseudo-Hermitian curvature model, para-Hermitian curvature model, hyper-pseudo-Hermitian curvature model, or hyper-para-Hermitian curvature model can be realized by a manifold with co…
We develop matrix models for Grassmann, flag, and Stiefel manifolds.
problem Creating efficient models for Grassmann, flag, and Stiefel manifolds.
method Orthogonally-equivariant matrix submanifold models derived for each manifold.
result Exhaustive list of orthogonally-equivariant submanifold models for the lowest dimensions.
Survey and clarify manifold-supported data in deep generative models.
problem Understanding why some DGMs succeed or fail at low-dimensional data.
method Formal analysis and new model connections.
result DGMs on autoencoder representations minimize Wasserstein distance.
Injective flows for star-like manifolds improve variational inference efficiency.
problem Efficiently modeling densities on star-like manifolds with exact Jacobian computation.
method Proposed injective flows for star-like manifolds with exact Jacobian computation.
result Exact Jacobian computation for star-like manifolds reduces computational cost to NFs.
New compact Weyl-parallel manifolds discovered in all dimensions n≥5.
problem Finding compact Weyl-parallel manifolds in all metric signatures and dimensions.
method Diffeomorphic to torus bundles over the circle, constructed from quotient-manifolds of model manifolds with discrete isometry groups.
result Existence of compact Weyl-parallel manifolds in all indefinite metric signatures in dimensions n≥5.
A new quantum gauge model is proposed. From this quantum gauge model we derive a quantum invariant of 3-manifolds. We show that this quantum invariant of 3-manifolds gives a classification of closed (orientable and connected) 3-manifolds. From this classification we then prove the Poincaré conjecture.
We define a combinatorial structure on 3-manifolds that combines the model manifolds constructed in Minsky's proof of the ending lamination conjecture with the layered triangulations defined by Jaco and Rubinstein.
Study shows how diffusion models learn on low-dimensional manifolds.
problem Learning efficiency of diffusion models on manifolds.
method Analyzes denoising score matching with random feature neural networks.
result Sample complexity scales linearly with intrinsic dimension, not ambient dimension.
This work proposes xGEMs or manifold guided exemplars, a framework to understand black-box classifier behavior by exploring the landscape of the underlying data manifold as data points cross decision boundaries. To do so, we train an unsupervised implicit generative model -- treated as a proxy to the data manifold. We …
We use splines and the Sasaki metric to analyze and compare manifold-valued trajectories.
problem Analyzing and comparing trajectories on Riemannian manifolds.
method Riemannian hierarchical model, Bézier splines, Sasaki metric.
result Spline-based approaches outperform state-of-the-art methods in intensity classification of trajectories.
SBMs learn manifold-like structures by mixing samples with a non-conservative field.
problem How SBMs learn data distributions on low-dimensional manifolds.
method Investigating linear approximations and subspaces of local feature vectors during diffusion.
result SBMs mix samples by a non-conservative field within the manifold, maintaining manifold-like structure.
Develops a new family of signature-changing models on metric manifolds.
problem Signature changes in metric manifolds.
method One-parameter family of Lorentz-Riemann models, local expressions around change.
result Generalizes existing signature-changing models.
Develops Shapley explainability solutions respecting data manifold.
problem Tenable assumption of uncorrelated features in Shapley explainability.
method Two solutions: generative modelling and direct learning of Shapley value-function.
result On-manifold Shapley explainability overcomes drawbacks of 'off-manifold' values.
The article generalizes Pearson correlation to Riemannian manifolds.
problem Analyzing statistical models on non-linear manifolds.
method Reconstitutes Pearson correlation properties and derives a nonlinear generalization.
result Developed the Riemann-Pearson Correlation for manifold analysis.
We consider the topic of multivariate regression on manifold-valued output, that is, for a multivariate observation, its output response lies on a manifold. Moreover, we propose a new regression model to deal with the presence of grossly corrupted manifold-valued responses, a bottleneck issue commonly encountered in pr…
Framework learns data manifold and generative model from corrupted data.
problem Learning from corrupted data with latent manifold structures.
method Riemannian AmbientFlow, incorporating normalizing flows and Riemannian Autoencoders.
result Framework recovers underlying data distribution and smooth manifold parametrization.
New algorithm proves most inflexible manifolds are not strongly inflexible.
problem Existence of simply-connected strongly inflexible manifolds.
method Algorithm based on Sullivan models.
result One example of simply-connected inflexible manifold is not strongly inflexible.
Quantum flag manifold σ-models are integrable and satisfy Ricci flow equations.
problem Integrating quantum flag manifold σ-models with fermions.
method Gauging bosonic Thirring/Gross-Neveu-type systems, adding fermions to cancel anomalies, and checking Ricci flow equations.
result Trigonometrically deformed geometries of flag manifold σ-models satisfy generalized Ricci flow equations.
Cannon and Swenson have shown that each hyperbolic 3-manifold group has a natural subdivision rule on the space at infinity, and that this subdivision rule captures the action of the group on the sphere. Explicit subdivision rules have also been found for some closed and finite-volume hyperbolic manifolds, as well as a…
New flows model distributions on Riemannian manifolds without domain knowledge.
problem Limited modeling of distributions on Riemannian manifolds.
method Riemannian convex potential maps using optimal transport.
result These flows can model standard distributions on spheres and tori.
In conventional Differential Geometry one studies manifolds, locally modelled on Rn, manifolds with boundary, locally modelled on [0,∞)×Rn−1, and manifolds with corners, locally modelled on [0,∞)k×Rn−k. They form categories ${\bf Man}\subset{\bf Man^b}\sub…
Bounded-type 3-manifolds arise as combinatorially bounded gluings of irreducible 3-manifolds chosen from a finite list. We prove effective hyperbolization and effective rigidity for a broad class of 3-manifolds of bounded type and large gluing heights. Specifically, we show the existence and uniqueness of hyperbolic me…
For k at least 2, we exhibit complete k-curvature homogeneous neutral signature pseudo-Riemannian manifolds which are not locally affine homogeneous (and hence not locally homogeneous). The curvature tensor of these manifolds is modeled on that of an indecomposible symmetric space. All the local scalar Weyl curvature i…
A method for learning distributions on complex manifolds using normalizing flows.
problem Learning distributions on non-Euclidean manifolds with high efficiency and accuracy.
method Learning a distribution on a manifold by combining local models that form an open cover.
result The method achieves better sample efficiency and competitive performance on manifolds of unknown topology.