arXiv research

Efficient trainable front-end for neural speech enhancement.

We shall investigate flat surfaces in hyperbolic 3-space with admissible singularities, called `flat fronts'. An Osserman-type inequality for complete flat fronts is shown. When equality holds in this inequality, we show that all the ends are embedded. Moreover, we shall give new examples for which equality holds.

It is classically known that complete flat surfaces in Euclidean 3-space are cylinders over space curves. This implies that the study of global behaviour of flat surfaces requires the study of singular points as well. If a flat surface $f$ admits singularities but its Gauss map $ν$ can be smoothly extended across the s…

FixyNN splits CNN models into fixed and trainable parts for efficient on-device inference.

The paper introduces a method to make neural networks more robust to adversarial attacks.

Sparsifying data reduces adversarial perturbation effects in deep networks.

It is by now well-known that small adversarial perturbations can induce classification errors in deep neural networks (DNNs). In this paper, we make the case that sparse representations of the input data are a crucial tool for combating such attacks. For linear classifiers, we show that a sparsifying front end is prova…

Novel CNN integrates learnable FIR filters for heart sound detection.

We propose a DTCWT ScatterNet Convolutional Neural Network (DTSCNN) formed by replacing the first few layers of a CNN network with a parametric log based DTCWT ScatterNet. The ScatterNet extracts edge based invariant representations that are used by the later layers of the CNN to learn high-level features. This improve…

WaveLSFormer learns profitable trading policies from financial time series data.

GPflow is a Gaussian process library that uses TensorFlow for its core computations and Python for its front end. The distinguishing features of GPflow are that it uses variational inference as the primary approximation method, provides concise code through the use of automatic differentiation, has been engineered with…

We provide complete source code for a front-end GUI and its back-end counterpart for a stock market visualization tool. It is built based on the "functional visualization" concept we discuss, whereby functionality is not sacrificed for fancy graphics. The GUI, among other things, displays a color-coded signal (computed…

We study a simple model of bicycle motion: a segment of fixed length in multi-dimensional Euclidean space, moving so that the velocity of the rear end is always aligned with the segment. If the front track is prescribed, the trajectory of the rear wheel is uniquely determined via a certain first order differential equa…

MTCNet uses MTL to estimate crowd density and count.

After Galvez, Martinez and Milan discovered a (Weierstrass-type) holomorphic representation formula for flat surfaces in hyperbolic 3-space, the first, third and fourth authors here gave a framework for complete flat fronts with singularities in H^3. In the present work we broaden the notion of completeness to weak com…

Short introduction to discrete flat fronts in hyperbolic space with a Weierstrass representation proof.

Wave fronts on certain surfaces become dense.

Study focal surfaces of wave fronts with unbounded curvatures.

The study classifies certain types of incomplete surfaces with low curvature.

Improved speaker diarization with attention-based embeddings.

The paper concerns a simple model of bicycle kinematics: a bicycle is represented by an oriented segment of constant length in n-dimensional space that can move in such a way that the velocity of its rear end is aligned with the segment (the rear wheel is fixed on the bicycle frame). Starting with a closed trajectory o…

New method finds exact Pareto front for MO-MDPs efficiently.

Bayesian optimization targets specific regions of the Pareto front in expensive multi-objective problems.

We characterize singularities of focal surfaces of wave fronts in terms of differential geometric properties of the initial wave fronts. Moreover, we study relationships between geometric properties of focal surfaces and geometric invariants of the initial wave fronts.

Study on singularities of Gauss maps of wave fronts with specific properties.

Cuspidal edges and swallowtails are typical non-degenerate singular points on wave fronts in the Euclidean $3$-space. Their first fundamental forms belong to a class of positive semi-definite metrics called "Kossowski metrics". A point where a Kossowski metric is not positive definite is called a singular point or a se…

RoyalFlush system improves multi-speaker ASR in M2MeT challenge.

The paper shows how null hypersurfaces behave in Lorentz-Minkowski space.

We simplify $D_4^+$-front singularities and apply to geometric invariants.

PFOPS extends PFO to multi-objective optimization.

New form of $D_4^-$-singularities for fronts in 3D space.

Study $D_4^-$-front singularities, compute invariants, and derive a Gauss-Bonnet theorem.

We give criteria for which a principal curvature becomes a bounded $C^\infty$-function at non-degenerate singular points of wave fronts by using geometric invariants. As applications, we study singularities of parallel surfaces and extended distance squared functions of wave fronts. Moreover, we relate these singularit…

We propose a design for schedule-based execution trading strategies based on uncertainty bands. This formulation: 1) simplifies strategy specification and implementation; 2) provides for flexible allocation among passive, opportunistic, aggressive, and dark pool crossing execution tactics; 3) allows for rapid enhanceme…

We propose a Lie geometric point of view on flat fronts in hyperbolic space as special omega-surfaces and discuss the Lie geometric deformation of flat fronts.

Proposes a new way to represent and analyze Pareto front surfaces.

FDR criterion simplifies complex causal graphs to a standard front-door setting.

New Bayesian method improves Pareto front estimation in multitask finetuning.

Novel NAS method balances performance and hardware metrics efficiently.

In this paper, we investigate the asymptotic behavior of regular ends of flat surfaces in the hyperbolic 3-space H^3. Galvez, Martinez and Milan showed that when the singular set does not accumulate at an end, the end is asymptotic to a rotationally symmetric flat surface. As a refinement of their result, we show that …

Recently Arnold's $\St$ and $J^{\pm}$ invariants of generic planar curves have been generalized to the case of generic planar wave fronts. We generalize these invariants to the case of wave fronts on an arbitrary surface $F$. All invariants satisfying the axioms which naturally generalize the axioms used by Arnold are …

This research focuses on optimizing expensive computer experiments with multiple objectives, targeting the central part of the Pareto front.

It is well-known that the unit cotangent bundle of any Riemannian manifold has a canonical contact structure. A surface in a Riemannian 3-manifold is called a (wave) front if it is the projection of a Legendrian immersion into the unit cotangent bundle. We shall give easily-computable criteria for a singular point on a…

A first order Vassiliev invariant of an oriented knot in an $S^1$-fibration and a Seifert fibration over a surface is constructed. It takes values in a quotient of the group ring of the first homology group of the total space of the fibration. It gives rise to an invariant of wave fronts on surfaces and orbifolds relat…

Two wave fronts $W_1$ and $W_2$ that originated at some points of the manifold $M^n$ are said to be causally related if one of them passed through the origin of the other before the other appeared. We define the causality relation invariant $CR (W_1, W_2)$ to be the algebraic number of times the earlier born front pass…

In this note we present a description of wave front evolving from an algebraic hypersurface by means of a pull-back of the discriminantal loci of a tame polynomial via a polynomial mapping. As an application we give examples of wave fronts which define free/almost free divisors near the focal point.

Study finds rigid submanifolds in spacelike waves under specific conditions.

We give a definition of `coherent tangent bundles', which is an intrinsic formulation of wave fronts. In our application of coherent tangent bundles for wave fronts, the first fundamental forms and the third fundamental forms are considered as induced metrics of certain homomorphisms between vector bundles. They satisf…