arXiv research

Paper uses SSL models' uncertainty to predict audio quality efficiently.

RUSLAN is a large Russian speech corpus for text-to-speech.

Developers of text-to-speech synthesizers (TTS) often make use of human raters to assess the quality of synthesized speech. We demonstrate that we can model human raters' mean opinion scores (MOS) of synthesized speech using a deep recurrent neural network whose inputs consist solely of a raw waveform. Our best models …

New algorithms optimize multiple machine learning metrics in real-world tasks.

MO-PaDGAN generates diverse, high-performance designs with multiple metrics.

Latent MoS learns multiple symmetries for efficient dynamic learning.

MO-PaDGAN improves multi-objective optimization by generating diverse and high-performing designs.

We propose a new methodology based on the Marshall-Olkin (MO) copula to model cross-border systemic risk. The proposed framework estimates the impact of the systematic and idiosyncratic components on systemic risk. Initially, we propose a maximum-likelihood method to estimate the parameter of the MO copula. In order to…

MO-GP models fill gaps in biophysical data with across-domain info transfer.

Despite the breakthroughs in accuracy and speed of single image super-resolution using faster and deeper convolutional neural networks, one central problem remains largely unsolved: how do we recover the finer texture details when we super-resolve at large upscaling factors? The behavior of optimization-based super-res…

We discuss a method to construct Dirac-harmonic maps developed by J.~Jost, X.~Mo and M.~Zhu in J.~Jost, X.~Mo, M.~Zhu, \emph{Some explicit constructions of Dirac-harmonic maps}, J. Geom. Phys. \textbf{59} (2009), no. 11, 1512--1527.The method uses harmonic spinors and twistor spinors, and mainly applies to Dirac-harmon…

New category theory for complex projective plane sections.

Sample-Rank simplifies MO recommendations by sampling and ranking, improving revenue with stable conversion rates.

For a fixed closed manifold $P$, we construct a cobordism category of embedded manifolds with a single Baas-Sullivan singularity of type $P$. Our main theorem identifies the homotopy type of the classifying space of this cobordism category with that of the infinite loop-space of a certain spectrum related to the spectr…

Paper proposes a transfer learning approach for decentralized QoE estimation.

A novel Bayesian optimization framework tackles multi-objective constrained problems.

USeMOC framework reduces expensive simulations for MO optimization with constraints.

Algorithm identifies Pareto set in bandits with contaminated feedback.

The notion of highly structured ring spectra of prime characteristic is made precise and is studied via the versal examples S//p for prime numbers p. These can be realized as Thom spectra, and therefore relate to other Thom spectra such as the unoriented bordism spectrum MO. We compute the Hochschild and André-Quillen …

MO-CBO optimizes multiple outcomes in causal systems with minimal data.

Discuss new policy learning objectives and methods.

Improved reinforcement learning with emergency stops.

Myopic optimization outperforms reinforcement learning in portfolio management, leading to lower returns and higher risks.

New algorithm learns and unlearns from streaming data efficiently.

A self-transverse immersion of a smooth manifold M^{k+2} in R^{2k+2} has a double point self-intersection set which is the image of an immersion of a smooth surface, the double point self-intersection surface. We prove that this surface may have odd Euler characteristic if and only if k is congruent to 1 modulo 4 or k+…

The paper analyzes Random Search and introduces BLiN-MOS for bandit learning.

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

Executing a basket of co-integrated assets is an important task facing investors. Here, we show how to do this accounting for the informational advantage gained from assets within and outside the basket, as well as for the permanent price impact of market orders (MOs) from all market participants, and the temporary imp…

We construct a pair of transverse genuine laminations on an atoroidal 3-manifold admitting transversely orientable uniform 1-cochain. The laminations are induced by the uniform 1-cochain and they are indeed the "straightening" of the coarse laminations defined in [Ca], by using minimal surface techniques. Moreover, whe…

User surveys for Quality of Experience (QoE) are a critical source of information. In addition to the common "star rating" used to estimate Mean Opinion Score (MOS), more detailed survey questions (problem tokens) about specific areas provide valuable insight into the factors impacting QoE. This paper explores two aspe…

We construct an extension of the Kontsevich integral of knots to knotted trivalent graphs, which commutes with orientation switches, edge deletions, edge unzips, and connected sums. In 1997 Murakami and Ohtsuki [MO] first constructed such an extension, building on Drinfel'd's theory of associators. We construct a step …

Recent progress in deep learning for audio synthesis opens the way to models that directly produce the waveform, shifting away from the traditional paradigm of relying on vocoders or MIDI synthesizers for speech or music generation. Despite their successes, current state-of-the-art neural audio synthesizers such as Wav…

Develops a new feature theory for robust machine learning.

The generalized Chen's conjecture on biharmonic submanifolds asserts that any biharmonic submanifold of a non-positively curved manifold is minimal (see e.g., [CMO1], [MO], [BMO1], [BMO2], [BMO3], [Ba1], [Ba2], [Ou1], [Ou2], [IIU]). In this paper, we prove that this conjecture is false by constructing foliations of pro…

In this paper we study unitary braid group representations associated with Majorana Fermions. Majorana Fermions are represented by Majorana operators, elements of a Clifford algebra. The paper recalls and proves a general result about braid group representations associated with Clifford algebras, and compares this resu…

Outlier detection is an important topic in machine learning and has been used in a wide range of applications. In this paper, we approach outlier detection as a binary-classification issue by sampling potential outliers from a uniform reference distribution. However, due to the sparsity of data in high-dimensional spac…

We give a new proof of the existence (\cite{HM}, \cite{Ren}) of a Jenkins-Strebel differential $Φ$ on a Riemann surface $\SR$ with prescribed heights of cylinders by considering the harmonic map from $\SR$ to the leaf space of the vertical foliation of $Φ$, thought of as a Riemannian graph. The novelty of the argument …

Using a combinatorial approach described in a recent paper of Manolescu, Ozsváth, and Sarkar we compute the Heegaard-Floer knot homology of all knots with at most 12 crossings as well as the $τ$ invariant for knots through 11 crossings. We review the basic construction of \cite{MOS}, giving two examples that can be wor…

We analyze the probability of ruin for the {\it scaled} classical Cramér-Lundberg (CL) risk process and the corresponding diffusion approximation. The scaling, introduced by Iglehart \cite{I1969} to the actuarial literature, amounts to multiplying the Poisson rate $\la$ by $n$, dividing the claim severity by $\sqrtn$, …

New framework integrates imitation and reinforcement learning for better robot performance.

Let $S_{g}$ denote the genus $g$ closed orientable surface. For $k\in \mathbb{N}$, a $k$-system is a collection of pairwise non-homotopic simple closed curves such that no two intersect more than $k$ times. Juvan-Malnič-Mohar \cite{Ju-Mal-Mo} showed that there exists a $k$-system on $S_{g}$ whose size is on the order o…

Conditional Value-at-Risk (CVaR) and Value-at-Risk (VaR), also called the superquantile and quantile, are frequently used to characterize the tails of probability distribution's and are popular measures of risk. Buffered Probability of Exceedance (bPOE) is a recently introduced characterization of the tail which is the…

A novel semi-supervised outlier detection model detects anomalies with few labels.

Enhances robustness of MOGP regression for multiple correlated outputs.

There are few known computable examples of non-abelian surface holonomy. In this paper, we give several examples whose structure 2-groups are covering 2-groups and show that the surface holonomies can be computed via a simple formula in terms of paths of 1-dimensional holonomies inspired by earlier work of Chan Hong-Mo…

New model handles missing data effectively in autoregressive models.

DiffWave generates high-fidelity audio waveforms efficiently.

MOMENT selects and estimates mixed-effects models using moment identities.