arXiv research

PRESTO improves rare event prediction by shrinking towards proportional odds model.

A new neural network model for ordinal regression.

The law of total probability may be deployed in binary classification exercises to estimate the unconditional class probabilities if the class proportions in the training set are not representative of the population class proportions. We argue that this is not a conceptually sound approach and suggest an alternative ba…

Bayesian model for discrete data with conditional transformations.

In this paper, we define a certain "proportional volume property" for an unit vector field on a spherical domain in S3. We prove that the volume of these vector fields has an absolute minimum and this value is equal to the volume of the Hopf vector field. Some examples of such vector fields are given. We also study the…

Paper introduces symmetric divergence link models for probability distributions.

Proposes a deep ordinal regression framework using optimal transport loss and unimodal output probabilities.

We find a unique torsion free Riemannian spin connection for the natural Killing metric on the quantum group $C_q[SL_2]$, using a recent frame bundle formulation. We find that its covariant Ricci curvature is essentially proportional to the metric (i.e. an Einstein space). We compute the Dirac operator and find for $q$…

Study non-linear Dirichlet-to-Neumann map for Poincaré-Einstein fillings.

Study proposes new methods to convert betting odds into accurate probabilities for sports forecasting.

Odd $K$-theory has the interesting property that it admits an infinite number of inequivalent differential refinements. In this paper we provide a bundle theoretic model for odd differential $K$-theory using the caloron correspondence and prove that this refinement is unique up to a unique natural isomorphism. We chara…

The paper derives theorems about curl eigenfields on a 3-sphere using angular momentum theory.

We consider odd Poisson (odd symplectic) structure on supermanifolds induced by an odd symmetric rank $2$ (non-degenerate) contravariant tensor field. We describe the difference between odd Riemannian and odd symplectic structure in terms of the Cartan prolongation of the corresponding Lie algebras, and formulate an an…

Paper improves deep learning for instance-level classification from label proportions.

If an artificial intelligence aims to maximise risk-adjusted return, then under mild conditions it is disproportionately likely to pick an unethical strategy unless the objective function allows sufficiently for this risk. Even if the proportion $η$ of available unethical strategies is small, the probability ${p_U}$ of…

Odd connections on supermanifolds are defined and their properties studied.

Proves a conjecture about knotted spheres using plane Floer homology.

Framework for fair predictive models using resampled sensitive attributes.

Modeling horse race betting odds with Ornstein-Uhlenbeck process.

We study the problem of learning with label proportions in which the training data is provided in groups and only the proportion of each class in each group is known. We propose a new method called proportion-SVM, or $\propto$SVM, which explicitly models the latent unknown instance labels together with the known group …

On any odd-dimensional oriented Riemannian manifold we define a volume form, which we call the odd Pfaffian, through a certain invariant polynomial with integral coefficients in the curvature tensor. We prove an intrinsic Chern-Gauss-Bonnet formula for incomplete edge singularities in terms of the odd Pfaffian on the f…

Develops a method to estimate average hazard under non-proportional hazards without relying on proportional hazards assumption.

We consider odd Laplace operators arising in odd symplectic geometry. Approach based on semidensities (densities of weight 1/2) is developed. The role of semidensities in the Batalin--Vilkovisky formalism is explained. In particular, we study the relations between semidensities on an odd symplectic supermanifold and di…

This paper extends NCFI to odd codimension and computes examples.

DSDE improves OoD detection by estimating model library proportions.

The divergence-like operator on an odd symplectic superspace which acts invariantly on a specially chosen odd vector field is considered. This operator is used to construct an odd invariant semidensity in a geometrically clear way. The formula for this semidensity is similar to the formula of the mean curvature of hype…

Federated Cox model handles non-proportional hazards in siloed data.

We establish several Witten type rigidity and vanishing theorems for twisted Toeplitz operators on odd dimensional manifolds. We obtain our results by combining the modular method, modular transgression and some careful analysis of odd Chern classes for cocycles in odd $K$-theory. Moreover we discover that in odd dimen…

In [D.A. Fedoseev, V.O. Manturov, A sliceness criterion for odd free knots,arXiv:1707.04923], the authors proved a sliceness criterion for odd free knots: free knots with odd chords. In the present paper we give a similar criterion for stably odd free knots. Some additional results on knot sliceness and cobordism are g…

Adapts scanning algorithm for odd Khovanov homology.

This paper refines previous work by the first author. We study the question of which links in the 3-sphere can be obtained as closures of a given 1-manifold in an unknotted solid torus in the 3-sphere (or genus-1 tangle) by adjoining another 1-manifold in the complementary solid torus. We distinguish between even and o…

learn2mix trains neural nets faster by adjusting class proportions dynamically.

Learning from Label Proportions (LLP) is a learning setting, where the training data is provided in groups, or "bags", and only the proportion of each class in each bag is known. The task is to learn a model to predict the class labels of the individual instances. LLP has broad applications in political science, market…

Framework simplifies vision-based control and goal discovery.

The present work investigates whether different quantification mechanisms (set comparison, vague quantification, and proportional estimation) can be jointly learned from visual scenes by a multi-task computational model. The motivation is that, in humans, these processes underlie the same cognitive, non-symbolic abilit…

New ODD metrics defined on manifolds with degeneracy conditions.

Odd-dimensional orbifolds' Euler characteristic equals half of their boundary's.

Paper extends index theorem to odd-dimensional manifolds with even-dimensional boundaries.

Proves another theorem for odd dimensional manifolds with boundary.

Proves module structure on odd Khovanov homology and applies to ribbon 2-knots.

New framework for weakly supervised learning from label proportions.

We consider odd Laplace operators acting on densities of various weight on an odd Poisson (= Schouten) manifold $M$. We prove that the case of densities of weight 1/2 (half-densities) is distinguished by the existence of a unique odd Laplace operator depending only on a point of an ``orbit space'' of volume forms. This…

Constructs modular forms and proves divisibility results for odd-dimensional manifolds.

Study odd generalized Einstein metrics on 3D Lie groups.

Extends odd Khovanov bracket to link cobordisms and proves functoriality up to sign.

StakeBench evaluates language understanding by linking comments to market commitments, improving model alignment with real-world outcomes.

We prove that an odd pretzel knot is doubly slice if it has $2n+1$ twist parameters consisting of $n+1$ copies of $a$ and $n$ copies of $-a$ for some odd integer $a$. Combined with the work of Issa and McCoy, it follows that these are the only doubly slice odd pretzel knots.

The moduli space of genus 3 translation surfaces with a single zero has two connected components. We show that in the odd connected component H^{odd}(4) the only GL^+(2,R) orbit closures are closed orbits, the Prym locus Q(3,-1^3), and H^{odd}(4). Together with work of Matheus-Wright, this implies that there are only f…