arXiv research

ResRep prunes CNNs without losing accuracy by separating remembering and forgetting.

The combination of the re-parameterization trick with the use of variational auto-encoders has caused a sensation in Bayesian deep learning, allowing the training of realistic generative models of images and has considerably increased our ability to use scalable latent variable models. The re-parameterization trick is …

Quantum advantage in derivative pricing requires 8k qubits and 54M T-depth.

This thesis presents some geometric insights into three different types of two player prediction games -- namely general learning task, prediction with expert advice, and online convex optimization. These games differ in the nature of the opponent (stochastic, adversarial, or intermediate), the order of the players' mo…

The paper proposes a new method for creating interpretable models using convex optimization.

New method samples Jeffreys prior for objective Bayesian inference.

The expectation-maximization (EM) algorithm is an iterative method for finding maximum likelihood estimates when data are incomplete or are treated as being incomplete. The EM algorithm and its variants are commonly used for parameter estimation in applications of mixture models for clustering and classification. This …

New method for Bayesian inference on large datasets.

Federated Learning is a distributed learning paradigm with two key challenges that differentiate it from traditional distributed optimization: (1) significant variability in terms of the systems characteristics on each device in the network (systems heterogeneity), and (2) non-identically distributed data across the ne…

The paper analyzes the amplitude of functions on the sphere, improving FDA methods.

Quaternion self-attention reduces computational cost and improves performance.

The goal of online prediction with expert advice is to find a decision strategy which will perform almost as well as the best expert in a given pool of experts, on any sequence of outcomes. This problem has been widely studied and $O(\sqrt{T})$ and $O(\log{T})$ regret bounds can be achieved for convex losses (\cite{zin…

New framework analyzes deep neural networks using feature probabilities.

AdaptOn achieves logarithmic regret in adaptive control of unknown partially observable linear systems.

Improved VI method for deep mixed models in finance.

Bayesian optimization improved for high-dimensional outputs using randomized priors.

Efficiently learns exponential family distributions with i.i.d. samples.

New method for initializing low-rank neural networks improves performance.

We simplify debiasing of LASSO estimates for practical computation.

The paper introduces a measure to assess the relative value of a delta-Symmetric Strangle under the Black-Scholes model.

ARMA nets expand receptive fields for dense prediction tasks.

Researchers develop a new spatial process model for non-Gaussian data.

We propose a new proximal, path-following framework for a class of constrained convex problems. We consider settings where the nonlinear---and possibly non-smooth---objective part is endowed with a proximity operator, and the constraint set is equipped with a self-concordant barrier. Our approach relies on the followin…

RAE improves image representation learning with simplified design choices.

The paper introduces new structures for left-symmetric algebroids.

We give a notion of compatibility between a Riemannian structure and a Jacobi structure. We prove that in case of fundamental examples of Jacobi structures : Poisson structures, contact structures and locally conformally symplectic structures, we get respectively Riemann-Poisson structures in the sense of M. Boucetta, …

We give a notion of compatibility between a Riemannian metric and a Jacobi structure. We prove that in case of Poisson structures, contact structures and locally conformally symplectic structures, fundamental examples of Jacobi structures, we get respectively Riemann-Poisson structures in the sense of M. Boucetta, $\fr…

Defines structure constants for specific geometric structures on Lie groups.

Study on $G_2^*$ structures and almost para-contact structures in 7D.

Defines a new Poisson structure for generalized Sasakian spaces.

In a preceding paper we introduced a notion of compatibility between a Jacobi structure and a Riemannian structure on a smooth manifold. We proved that in the case of fundamental examples of Jacobi structures : Poisson structures, contact structures and locally conformally symplectic structures, we get respectively Rie…

Study on types of generalized hypercomplex structures on tori and Kodaira-Thurston surface.

Classifies complex Dirac structures with invariants and local structure.

Extends corner structure study to general case, constructs normal Trans-Sasakian structures.

The paper explores geometric structures on Hom-Lie groups and algebras.

New metric structures generalize Sasakian and cosymplectic structures, proving rigidity and finding conditions.

3D projective structures can be metrized with conformal structures.

Hypersymplectic structures with torsion on Lie algebroids are investigated. We show that each hypersymplectic structure with torsion on a Lie algebroid determines three Nijenhuis morphisms. From a contravariant point of view, these structures are twisted Poisson structures. We prove the existence of a one-to-one corres…

Study equivalence between Hessian and Born structures on tangent bundles.

Introduces compatibility between Dirac structures and Nijenhuis tensors.

We introduce generalized almost contact structures which admit the $B$-field transformations on odd dimensional manifolds. We provide definition of generalized Sasakain structures from the view point of the generalized almost contact structures. We obtain a generalized Sasakian structure on a non-compact manifold which…

Introduces semi-abelian generalized complex structures.

Introduces VB-structures for geometric objects on manifolds.

The paper studies nilpotent structures in oriented neutral vector bundles and neutral hyperKähler structures.

Spin-structures on real Bott manifolds with Kähler structure are characterized.

Expanding on previous work, this note generalizes geometric structures results.

Exotic hypercomplex structures on a torus are proven to not exist.

Parabolic almost conformally symplectic structures were introduced in the first part of this series of articles as a class of geometric structures which have an underlying almost conformally symplectic structure. If this underlying structure is conformally symplectic, then one obtains a PCS-structure. In the current ar…