arXiv research

Dual representations for robust risk measures and uncertainty sets.

Drago optimizes DRO problems with faster convergence.

We define a class of L-convex-concave subsets of $\Bbb{R}P^n$, where L is a projective subspace of dimension l in $\Bbb{R}P^n$. These are sets whose sections by any (l+1)-dimensional space L' containing L are convex and concavely depend on L'. We introduce an L-duality for these sets, and prove that the L-dual to an L-…

Paper explores closedness properties of convex sets in rearrangement invariant spaces.

Established a new inequality for convex bodies in high dimensions.

A new method for distributed optimization reduces communication rounds without minibatches.

Unified algorithm solves convex optimization problems with optimal rates.

Extends dual volume and curvature measures to broader functions and sets, solving Minkowski problems.

The paper extends NUP representations to factor graphs for better estimation.

We focus on the robust principal component analysis (RPCA) problem, and review a range of old and new convex formulations for the problem and its variants. We then review dual smoothing and level set techniques in convex optimization, present several novel theoretical results, and apply the techniques on the RPCA probl…

New divergence defined in Information Geometry mimics Riemannian geodesics.

Iterative Hard Thresholding (IHT) is a class of projected gradient descent methods for optimizing sparsity-constrained minimization models, with the best known efficiency and scalability in practice. As far as we know, the existing IHT-style methods are designed for sparse minimization in primal form. It remains open t…

Paper solves a complex equation for unbounded convex sets.

We consider empirical risk minimization of linear predictors with convex loss functions. Such problems can be reformulated as convex-concave saddle point problems, and thus are well suitable for primal-dual first-order algorithms. However, primal-dual algorithms often require explicit strongly convex regularization in …

Let $(Φ,Ψ)$ be a conjugate pair of Orlicz functions. A set in the Orlicz space $L^Φ$ is said to be order closed if it is closed with respect to dominated convergence of sequences of functions. A well known problem arising from the theory of risk measures in financial mathematics asks whether order closedness of a conve…

In approachability with full monitoring there are two types of conditions that are known to be equivalent for convex sets: a primal and a dual condition. The primal one is of the form: a set C is approachable if and only all containing half-spaces are approachable in the one-shot game; while the dual one is of the form…

Improves SDCA convergence for convex objectives with linear constraints.

Study shows infimum of dual volume equals convex core volume for hyperbolic 3-manifolds.

Accelerates machine learning algorithms for sparse data.

Hadwiger's Theorem states that Euclidean-invariant convex-continuous valuations of definable sets are linear combinations of intrinsic volumes. We lift this result from sets to data distributions over sets, specifically, to definable real-valued functions on n-dimensional Euclidean space. This generalizes intrinsic vol…

Researchers prove uniqueness and continuity of solution to L_p dual Minkowski problem.

Dual explanation method using convex hulls and example-based vectors.

Set risk measures extend traditional risk measures to handle sets of positions.

Equivalent characterizations of multiportfolio time consistency are deduced for closed convex and coherent set-valued risk measures on $L^p(Ω,\mathcal F, P; R^d)$ with image space in the power set of $L^p(Ω,\mathcal F_t,P;R^d)$. In the convex case, multiportfolio time consistency is equivalent to a cocycle condition on…

In this paper, the following three are shown. (1) For a $C^\infty$ convex integrand $γ: S^n\to \mathbb{R}_+$, its dual convex integrand $δ: S^n\to \mathbb{R}_+$ is of class $C^\infty$. (2) For a stable convex integrand $γ: S^n\to \mathbb{R}_+$, its dual convex integrand $δ: S^n\to \mathbb{R}_+$ is stable. (3) Let $γ: S…

Dual-based algorithms optimize distributed convex problems over networks.

In this paper, we investigate simultaneous properties of a convex integrand $γ$ and its dual $δ$. The main results are the following three. (1) For a $C^\infty$ convex integrand $γ: S^n\to \mathbb{R}_+$, its dual convex integrand $δ: S^n\to \mathbb{R}_+$ is of class $C^\infty$ if and only if $γ$ is a strictly convex in…

This work studies the strong duality of non-convex matrix factorization problems: we show that under certain dual conditions, these problems and its dual have the same optimum. This has been well understood for convex optimization, but little was known for non-convex problems. We propose a novel analytical framework an…

The classical duality theory of Kantorovich and Kellerer for the classical optimal transport is generalized to an abstract framework and a characterization of the dual elements is provided. This abstract generalization is set in a Banach lattice $\cal{X}$ with a order unit. The primal problem is given as the supremum o…

PURE-CD algorithm proves complexity bounds for convex-concave problems.

Develops risk measures for markets with constraints and costs.

In this note we introduce a natural Finsler structure on convex surfaces, referred to as the projective Finsler structure, which is dual in a sense to the obvious inclusion of a convex surface in a normed space. It has an associated projective girth, which is similar to the notion of girth defined by Schäffer. We prove…

New algorithms solve convex-concave problems faster than previous methods.

We consider the convex-concave saddle point problem $\min_{x}\max_{y} f(x)+y^\top A x-g(y)$ where $f$ is smooth and convex and $g$ is smooth and strongly convex. We prove that if the coupling matrix $A$ has full column rank, the vanilla primal-dual gradient method can achieve linear convergence even if $f$ is not stron…

Smooth even solutions found for a generalized convex geometry problem.

Dual volume bound linked to Weil-Petersson distance in quasi-Fuchsian manifolds.

The dual Minkowski problem for even data asks what are the necessary and sufficient conditions on an even prescribed measure on the unit sphere for it to be the $q$-th dual curvature measure of an origin-symmetric convex body in $\mathbb{R}^n$. A full solution to this is given when $1 < q < n$. The necessary and suffic…

Convex dual network improves neural network reconstruction for medical imaging.

The paper optimizes training samples for image denoising across different noise levels.

Solves a generalized dual Minkowski problem for specific values of q.

Develops a new theory of loss functions for statistical machine learning.

Study on Santaló point for convex bodies in normed spaces.

Unified framework for robust risk measures beyond convexity.

Dual-ISL improves implicit generative model training with convex optimization and explicit density approximation.

We provide a dual representation of quasiconvex maps between two lattices of random variables in terms of conditional expectations. This generalizes the dual representation of quasiconvex real valued functions and the dual representation of conditional convex maps.

PDA method optimizes neural networks with global convergence rate analysis.

Extends convex clustering to graph-structured data.

We provide a characterization in terms of Fatou closedness for weakly closed monotone convex sets in the space of $\mathcal{P}$-quasisure bounded random variables, where $\mathcal{P}$ is a (possibly non-dominated) class of probability measures. Applications of our results lie within robust versions the Fundamental Theo…