arXiv research

Proposes a self-supervised method for generating spatial audio from monaural audio and video.

We investigate variations of Brieskorn lattices over non-compact parameter spaces, and discuss the corresponding limit objects on the boundary divisor. We study the associated variation of twistors and the corresponding limit mixed twistor structures. We construct a compact classifying space for regular singular Briesk…

We determine the extent to which the collection of $Γ$-Euler-Satake characteristics classify closed 2-orbifolds. In particular, we show that the closed, connected, effective, orientable 2-orbifolds are classified by the collection of $Γ$-Euler-Satake characteristics corresponding to free or free abelian $Γ$ and are not…

We show that, for generative classifiers, conditional independence corresponds to linear constraints for the induced discrimination functions. Discrimination functions of undirected Markov network classifiers can thus be characterized by sets of linear constraints. These constraints are represented by a second order fi…

Classifies polar actions on 3D homogeneous spaces.

We classify all of the 4-dimensional linear Poisson structures of which the corresponding Lie algebras can be considered as the extension by a derivation of 3-dimensional unimodular Lie algebras. The affine Poisson structures on R^3 are totally classified.

The study classifies constant mean curvature surfaces in curved spaces.

J. Boyle classified 1-handles attached to surface-knots, that are closed and connected surfaces embedded in the Euclidean 4-space, in the case that the surfaces are oriented and 1-handles are orientable with respect to the orientations of the surfaces. In that case, the equivalence classes of 1-handles correspond to th…

The paper tackles binary classification with measure data using topological descriptors.

Discriminative classifiers improve decision-making in SHM systems.

Study controls error rates of binary classifiers using hypothesis testing.

A new method combines simple binary classifiers to build complex multiclass classifiers, achieving performance limits in a Gaussian setting.

Currently, engineers at substation service providers match customer data with the corresponding internally used signal names manually. This paper proposes a machine learning method to automate this process based on substation signal mapping data from a repository of executed projects. To this end, a bagged token classi…

This paper considers the problem of removing costly features from a Bayesian network classifier. We want the classifier to be robust to these changes, and maintain its classification behavior. To this end, we propose a closeness metric between Bayesian classifiers, called the expected classification agreement (ECA). Ou…

Valence two Killing tensors in the Euclidean and Minkowski planes are classified under the action of the group which preserves the type of the corresponding Killing web. The classification is based on an analysis of the system of determining partial differential equations for the group invariants and is entirely algebr…

Classifies totally geodesic submanifolds and polar actions on Stiefel manifolds.

We classify Legendrian knots of topological type $7_6$ having maximal Thurston--Bennequin number confirming the corresponding conjectures of Chongchitmate--Ng.

Classifies singularities in quiver varieties for specific Dynkin quivers.

We classify real 6-dimensional nilpotent Lie algebras for which the corresponding Lie group has a left-invariant complex structure, and estimate the dimensions of moduli spaces of such structures.

We explore the question of whether the representations learned by classifiers can be used to enhance the quality of generative models. Our conjecture is that labels correspond to characteristics of natural data which are most salient to humans: identity in faces, objects in images, and utterances in speech. We propose …

We study generalized Kaehler manifolds for which the corresponding complex structures commute and classify completely the compact generalized Kaehler four-manifolds for which the induced complex structures yield opposite orientations.

The Nearest subspace classifier (NSS) finds an estimation of the underlying subspace within each class and assigns data points to the class that corresponds to its nearest subspace. This paper mainly studies how well NSS can be generalized to new samples. It is proved that NSS is strongly consistent under certain assum…

Classifies toric dually flat manifolds into complex space forms.

In acoustic scene classification researches, audio segment is usually split into multiple samples. Majority voting is then utilized to ensemble the results of the samples. In this paper, we propose a punishment voting algorithm based on the super categories construction method for acoustic scene classification. Specifi…

A qualgebra $G$ is a set having two binary operations that satisfy compatibility conditions which are modeled upon a group under conjugation and multiplication. We develop a homology theory for qualgebras and describe a classifying space for it. This space is constructed from $G$-colored prisms (products of simplices) …

Classifies real trivectors in 9D, following complex classification methods.

The infinite matrix `Schwartz' group $G^{-\infty}$ is a classifying group for odd K-theory and carries Chern classes in each odd dimension, generating the cohomology. These classes are closely related to the Fredholm determinant on $G^{-\infty}.$ We show that while the higher (even, Schwartz) loop groups of $G^{-\infty…

The main goal is to classify 4-dimensional real Lie algebras $\g$ which admit a para-hypercomplex structure. This is a step toward the classification of Lie groups admitting the corresponding left-invariant structure and therefore possessing a neutral, left-invariant, anti-self-dual metric. Our study is related to the …

Adaptive classifier optimizes high-dimensional data with spiked covariance structure.

The paper uses geometric methods to classify medical data histograms.

New method controls classifier guidance in diffusion models.

Generalizes classifying spaces for topological groups with torsion.

We classify the simple sheaves microsupported along the conormal bundle of a knot. We also establish a correspondence between simple sheaves up to local systems and augmentations, explaining the underlying reason why knot contact homology representations detect augmentations.

The aim of this paper is to classify compact, simply connected Kähler manifolds which admit J-invariant Killing tensor with two eigenvalues of multiplicity 2 and n-2 and with constant eigenvalue corresponding to 2-dimensional eigendistribution.

Efficient classifier error estimation without re-training.

The paper classifies symmetric triads with multiplicities and their applications.

Classifies worst approximable rational numbers using hyperbolic geometry.

A method for learning discontinuous functions using clustering, classification, and regression.

Paper presents a method to extract and interpret knowledge from a spiking neural classifier.

Randomised classifiers outperform deterministic ones in strategic classification.

Sparse nearest-centroid classifiers detect relevant features for classification.

The variational autoencoder (VAE) is a popular probabilistic generative model. However, one shortcoming of VAEs is that the latent variables cannot be discrete, which makes it difficult to generate data from different modes of a distribution. Here, we propose an extension of the VAE framework that incorporates a classi…

The paper classifies orbits of semisimple elements in real semisimple Lie algebras.

Classifies non-integrable distributions with simple infinite-dimensional Lie superalgebras of symmetries.

Study classifies Einstein-Yang-Mills spaces in 4D symmetric spaces.

New MC simulation methods use classifiers to estimate pdf ratios without explicit pdfs.

Study extended Bogomolny equations on curved space with special boundary conditions.

In this paper, we propose a novel dynamic ensemble selection framework using meta-learning. The framework is divided into three steps. In the first step, the pool of classifiers is generated from the training data. The second phase is responsible to extract the meta-features and train the meta-classifier. Five distinct…