arXiv research

The paper evaluates machine learning for song similarity based on metadata and user tags.

This paper investigates methods for quantifying similarity between audio signals, specifically for the task of of cover song detection. We consider an information-theoretic approach, where we compute pairwise measures of predictability between time series. We compare discrete-valued approaches operating on quantised au…

The use of community detection algorithms is explored within the framework of cover song identification, i.e. the automatic detection of different audio renditions of the same underlying musical piece. Until now, this task has been posed as a typical query-by-example task, where one submits a query song and the system …

SCM-GAN converts any song to sound like a different singer.

Paper classifies Brazilian music genres using song lyrics with BLSTM network.

This work formulates a novel song recommender system as a matrix completion problem that benefits from collaborative filtering through Non-negative Matrix Factorization (NMF) and content-based filtering via total variation (TV) on graphs. The graphs encode both playlist proximity information and song similarity, using …

Predicts which songs will be Billboard hits using Spotify data.

Improved cover song detection with neural networks.

Model predicts hit songs in pop music using Siamese CNN with ranking loss.

Uniform estimates for Kaehler metrics' diameters and volumes.

Predicting dance hits from 1985-2013 using musical features.

Deep learning models predict hit songs from audio features.

SeER hybrid model improves song recommendations and explains them.

We conjecture that the Joyce-Song wall-crossing formula for Donaldson-Thomas invariants arises naturally from an asymptotic expansion in the field theoretic work of Gaiotto, Moore and Neitzke. This would also give a new perspective on how the formulae of Joyce-Song and Kontsevich-Soibelman are related. We check the con…

A method for cold-start playlist recommendation using multitask learning.

We find the minimum dilatation of pseudo-Anosov braids on n-punctured discs for 3 <= n <= 8. This covers the results of Song-Ko-Los (n=4) and Ham-Song (n=5). The proof is elementary, and uses the Lefschetz formula.

Model learns to sort music clips in sequence.

The study uses historical revenue data to forecast music catalog cashflows and multipliers.

Framework generates pop song melodies and piano accompaniment.

Improved neural network training by coupled initialization reduces neuron count.

Minimal surfaces with negative curvature found in large spheres.

Neural nets classify Thai Lukthung songs from other genres.

Rating prediction is an important application, and a popular research topic in collaborative filtering. However, both the validity of learning algorithms, and the validity of standard testing procedures rest on the assumption that missing ratings are missing at random (MAR). In this paper we present the results of a us…

Study bounds Kähler current diameters on manifolds.

Estimates the growth of Morse index for free boundary minimal hypersurfaces.

In this paper, we give a criterion for the properness of the K-energy in a general Kahler class of a compact Kahler manifold by using Song-Weinkove's result. As applications, we give some Kahler classes on $\mathbb{C}\mathbb{P}^2\#3\overline {\mathbb{C}\mathbb{P}^2}$ and $\mathbb{C}\mathbb{P}^2\#8\overline {\mathbb{C}\…

Cai, Song and Kou (2015) [Cai, N., Y. Song, S. Kou (2015) A general framework for pricing Asian options under Markov processes. Oper. Res. 63(3): 540-554] made a breakthrough by proposing a general framework for pricing both discretely and continuously monitored Asian options under one-dimensional Markov processes. In …

CantorNet tests geometric and topological complexity in neural networks.

The paper proves infinitely many free boundary minimal hypersurfaces in compact manifolds.

The abstract proves there are infinitely many minimal surfaces in 3-7 dimensions.

Jukebox generates high-fidelity songs with singing in raw audio.

The paper solves a 5-manifold foliation problem using a Sasaki-Ricci flow.

Which song will Smith listen to next? Which restaurant will Alice go to tomorrow? Which product will John click next? These applications have in common the prediction of user trajectories that are in a constant state of flux over a hidden network (e.g. website links, geographic location). What users are doing now may b…

We give examples of smooth surfaces with negative first Chern class which are slope unstable with respect to certain polarisations, and so have Kahler classes that do not admit any constant scalar curvature Kahler metrics. We also compare this to the work of Song-Weinkove on the J-flow.

This note explains Ricci flow method for Kähler-Einstein metrics.

Researchers prove a Penrose inequality for spacetime with specific conditions.

Paper evaluates word embedding techniques on Russian digital humanities datasets.

In this paper, we give a result on the properness of the K-energy, which answers a question of Song-Weinkove in any dimensions. Moreover, we extend our previous result on the properness of K-energy to the case of modified K-energy associated to extremal Kahler metrics.

Paper proposes singing voice conversion without parallel data.

We generalize the maximal time existence of Kähler-Ricci flow in Tian-Zhang and Song-Tian to conical case. Furthermore, if the twisted canonical bundle $K_{M}+(1-β)[D]$ is big or big and nef, we can expect more on the limit behaviors of such conical Kähler-Ricci flow. Moreover, the results still hold for simple normal …

Let $X = M \times E$ where $M$ is an $m$-dimensional Kähler manifold with negative first Chern class and $E$ is an $n$-dimensional complex torus. We obtain $C^\infty$ convergence of the normalized Kähler-Ricci flow on $X$ to a Kähler-Einstein metric on $M$. This strengthens a convergence result of Song-Weinkove and con…

A crossing in a knot is nugatory if changing the crossing does not change the knot type. Using an invariant of certain types of closed 3-braid diagrams, we show that if a closed 3-braid contains a nugatory crossing then its braid index is one or two. This proves a special case of a conjecture on nugatory crossings due …

The paper analyzes skewness and kurtosis measures for skew-elliptical distributions.

Given a matrix $A\in \mathbb{R}^{n\times d}$ and a vector $b\in \mathbb{R}^n$, we consider the regression problem with $\ell_\infty$ guarantees: finding a vector $x'\in \mathbb{R}^d$ such that $\|x'-x^*\|_\infty \leq \fracε{\sqrt{d}}\cdot \|Ax^*-b\|_2\cdot \|A^\dagger\|$ where $x^*=\arg\min_{x\in \mathbb{R}^d}\|Ax-b\|…

Flow analysis leads to metric completion in Kähler geometry.

We introduce a new continuity method which provides an alternative way of carrying out the Analytic Minimal Model Program introduced by G. Tian and J. Song and G. Tian. This equation -- unlike the Ricci flow -- has the advantage of having Ricci curvature bounded from below along the deformation, so that the compactness…

System composes polyphonic music using LSTM and RL.

We investigate the Kahler-Ricci flow on holomorphic fiber spaces whose generic fiber is a Calabi-Yau manifold. We establish uniform metric convergence to a metric on the base, away from the singular fibers, and show that the rescaled metrics on the fibers converge to Ricci-flat Kahler metrics. This strengthens previous…