Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,657 papers · 148 categories

Trend · papers per month

3673109145 · Jun 202019922001200920172026
48 results for imbalanced length pairs

Meta-learning framework improves short utterance speaker recognition.

problem Poor performance of existing models with short utterances.
method Prototypical Networks with support and query sets, enforcing classification against entire training set.
result Significant performance gains on VoxCeleb datasets.

Paper estimates optimal ROC curve arc length and AUC, improving classification performance.

problem Estimating optimal ROC curve arc length and AUC in imbalanced binary classification.
method Expresses arc length and AUC as variational objectives, estimating using positive and negative samples.
result Proposed classification procedure maximizes an approximate lower bound of maximal AUC.

Study on the minimum length of curves on once-punctured hyperbolic surfaces.

problem Finding the minimum length of filling pairs on once-punctured hyperbolic surfaces.
method Analyzing the topology and geometry of the surface to derive a lower bound for the length of filling pairs.
result A lower bound for the length of filling pairs on once-punctured hyperbolic surfaces is derived, depending only on the surface's topology.

This paper classifies and determines the length of the shortest filling pairs on a specific type of surface.

problem Classifying and determining the length of the shortest filling pairs on a specific type of surface.
method Classifying and determining the length of the shortest filling pairs on a specific type of surface.
result The paper classifies and determines the length of the shortest minimal filling pairs on a genus two surface.

We study the spectral properties of a large class of compact flat Riemannian manifolds of dimension 4, namely, those whose corresponding Bieberbach groups have the canonical lattice as translation lattice. By using the explicit expression of the heat trace of the Laplacian acting on pp-forms, we determine all pp-isos…

2005-05-23abs ↗pdf ↗

Given a compact orientable surface of negative Euler characteristic, there exists a natural pairing between the Teichmueuller space of the surface and the set of homotopy classes of simple loops and arcs. The length pairing sends a hyperbolic metric and a homotopy class of a simple loop or arc to the length of geodesic…

2002-11-27abs ↗pdf ↗

A pair (α,β)(α, β) of simple closed geodesics on a closed and oriented hyperbolic surface MgM_g of genus gg is called a filling pair if the complementary components of αβα\cupβ in MgM_g are simply connected. The length of a filling pair is defined to be the sum of their individual lengths. In \cite{Aou}, Aougab-Huang con…

2019-07-16abs ↗pdf ↗

We study the length, weak length and complex length spectrum of closed geodesics of a compact flat Riemannian manifold, comparing length-isospectrality with isospectrality of the Laplacian acting on p-forms. Using integral roots of the Krawtchouk polynomials, we give many pairs of p-isospectral flat manifolds having di…

2001-10-31abs ↗pdf ↗

A pair of distinct free homotopy classes of closed curves in an orientable surface FF with negative Euler characteristic is said to be length equivalent if for any hyperbolic structure on FF, the length of the geodesic representative of one class is equal to the length of the geodesic representative of the other clas…

2015-11-20abs ↗pdf ↗

This paper extends neural collapse to class-imbalanced datasets using an unconstrained ReLU feature model.

problem Understanding neural collapse in class-imbalanced datasets with cross-entropy loss.
method Generalized neural collapse to class-imbalanced settings using an unconstrained ReLU feature model.
result Class-means converge to orthogonal vectors with different lengths, and classifier weights align to these vectors.

In this paper, we establish upper bounds on the length of the shortest conjugator between pairs of infinite order elements in a wide class of groups. We obtain a general result which applies to all hierarchically hyperbolic groups, a class which includes mapping class groups, right-angled Artin groups, Burger--Mozes-ty…

2018-08-29abs ↗pdf ↗

Deep linear networks exhibit collapsing features and classifiers across datasets.

problem Understanding the collapse of features and classifiers in deep linear networks.
method Theoretical and empirical analysis of deep linear networks with MSE and CE losses.
result Deep linear networks exhibit NC properties, collapsing features and classifiers to orthogonal vectors.

A common issue for classification in scientific research and industry is the existence of imbalanced classes. When sample sizes of different classes are imbalanced in training data, naively implementing a classification method often leads to unsatisfactory prediction results on test data. Multiple resampling techniques…

2020-02-11abs ↗pdf ↗

Study mixed commutator lengths in wreath products and their relation to general ranks.

problem Understanding mixed commutator lengths in wreath products and their relation to general ranks.
method Analyzing wreath products (G,N)=(ZΓ,ΓZ)(G,N)=(\mathbb{Z}\wr Γ, \bigoplus_Γ\mathbb{Z}) and determining mixed commutator lengths in terms of general rank.
result Mixed commutator lengths and ordinary commutator lengths coincide under certain conditions.

Study of group actions on CAT(0) cube complexes, focusing on marked length spectra.

problem Comparing marked length spectra of group actions on CAT(0) cube complexes.
method Use of finite-state automata and thermodynamic formalism for suspension flows over subshifts of finite type.
result Prove that the Manhattan curve is analytic and convex, and a straight line if and only if marked length spectra are homothetic.

Two Riemannian manifolds are called eigenvalue equivalent when their sets of eigenvalues of the Laplace-Beltrami operator are equal (ignoring multiplicities). They are (primitive) length equivalent when the sets of lengths of their (primitive) closed geodesics are equal. We give a general construction of eigenvalue equ…

2006-06-14abs ↗pdf ↗

Reid has asked whether hyperbolic manifolds with the same geodesic length spectrum must be commensurable. Building toward a negative answer to this question, we construct examples of hyperbolic 3-manifolds that share an arbitrarily large portion of the length spectrum but are not commensurable. More precisely, for all …

2016-09-02abs ↗pdf ↗

Study on combinatorial kk-systoles on surfaces, showing growth in intersection numbers.

problem Understanding the intersection numbers of closed curves on surfaces.
method Analyzing combinatorial kk-systoles on punctured tori and pairs of pants.
result The maximal intersection number of combinatorial kk-systoles grows like kk and approaches infinity as kk increases.

It is well-known that any pair of closed orientable 3-manifolds are related by a finite sequence of Dehn surgeries on knots. Furthermore Kawauchi showed that such knots can be taken to be hyperbolic. In this article, we consider the minimal length of such sequences connecting a pair of 3-manifolds, in particular, a pai…

2008-09-22abs ↗pdf ↗

Thurston introduced shear deformations (cataclysms) on geodesic laminations - deformations including left and right displacements along geodesics. For hyperbolic surfaces with cusps, we consider shear deformations on disjoint unions of ideal geodesics. The length of a balanced weighted sum of ideal geodesics is defined…

2013-03-01abs ↗pdf ↗

Two free homotopy classes of closed curves in an orientable surface with negative Euler characteristic are said to be length equivalent if for any hyperbolic structure on the surface, the length of the geodesic in one class is equal to the length of the geodesic in the other class. We show that there are elements in th…

2013-11-03abs ↗pdf ↗

New proof shows surfaces can have identical length spectra but not simple ones.

problem Identifying when two covers of a surface have identical length spectra but not simple ones.
method Characterized isomorphism of covers and constructed surfaces with identical spectra but different simple length spectra.
result Found surfaces with identical length spectra but not simple length isospectral covers.

VIR model improves regression accuracy and uncertainty estimation for imbalanced data.

problem Imbalanced regression datasets lead to poor model accuracy and uncertainty estimation.
method VIR model uses probabilistic smoothing and reweighting to estimate latent representations and uncertainty.
result VIR model outperforms state-of-the-art models in accuracy and uncertainty estimation.

Classifiers trained on data sets possessing an imbalanced class distribution are known to exhibit poor generalisation performance. This is known as the imbalanced learning problem. The problem becomes particularly acute when we consider incremental classifiers operating on imbalanced data streams, especially when the l…

2014-05-09abs ↗pdf ↗

Study shows imbalanced labels can be beneficial but not always in class-imbalanced learning.

problem Challenges in class-imbalanced learning with heavy label bias.
method Systematic investigation of semi-supervised and self-supervised approaches to leverage imbalanced labels.
result Imbalanced labels are valuable in semi-supervised learning but not always in self-supervised learning.

We introduce a pair of isospectral but non-isometric compact flat 3-manifolds called Tetra (a tetracosm) and Didi (a didicosm). The closed geodesics of Tetra and Didi are very different. Where Tetra has two quarter-twisting geodesics of the shortest length, Didi has four half-twisting geodesics. Nevertheless, these spa…

2004-07-25abs ↗pdf ↗

Optimal downsampling improves GLM performance in imbalanced classification.

problem Improving GLM performance in imbalanced classification.
method Proposed a pseudo maximum likelihood estimator for optimal downsampling.
result The introduced estimator outperforms existing alternatives in both synthetic and empirical data.

Proposes a new data augmentation method for imbalanced datasets in both classification and regression.

problem Imbalanced datasets in supervised learning, especially in regression.
method GOLIATH algorithm based on kernel density estimates for classification and regression.
result Significant improvement over existing state-of-the-art techniques in imbalanced regression.