arXiv research

Proposes a novel tensor-based approach for multi-level link prediction.

Study on exotic smooth embeddings of surfaces in 4-manifolds, revealing different properties and complexities.

The paper introduces heterogeneous manifolds for better graph embeddings.

Develops a new causal model for path-dependent link prediction.

Engel structures on bundles over 3-manifolds in complex 3-space.

New theorem on embedding Moebius bands in 3D space.

We investigate the use of Minimax distances to extract in a nonparametric way the features that capture the unknown underlying patterns and structures in the data. We develop a general-purpose and computationally efficient framework to employ Minimax distances with many machine learning methods that perform on numerica…

New method preserves distances in time series data.

Recent work in learning ontologies (hierarchical and partially-ordered structures) has leveraged the intrinsic geometry of spaces of learned representations to make predictions that automatically obey complex structural constraints. We explore two extensions of one such model, the order-embedding model for hierarchical…

Minimal hypertori found in 4D sphere, solving Bernstein conjecture.

Sampling a fraction of pairs can match full evaluation in machine learning losses.

AutoEmbedder clusters unlabeled data using semi-supervised DNN embedding.

Study shows gMPNNs struggle with OOD link prediction in larger test graphs.

LOT Wassmap speeds up Wasserstein space manifold learning.

Knowledge graphs are a versatile framework to encode richly structured data relationships, but it can be challenging to combine these graphs with unstructured data. Methods for retrofitting pre-trained entity representations to the structure of a knowledge graph typically assume that entities are embedded in a connecte…

Recently, graph neural networks have attracted great attention and achieved prominent performance in various research fields. Most of those algorithms have assumed pairwise relationships of objects of interest. However, in many real applications, the relationships between objects are in higher-order, beyond a pairwise …

New method models aptamer libraries as Boltzmann-weighted graph ensembles for better affinity predictions.

Study metric learning from limited preference comparisons, showing how low-dimensional structure can still reveal metric information.

This paper presents a distance-based discriminative framework for learning with probability distributions. Instead of using kernel mean embeddings or generalized radial basis kernels, we introduce embeddings based on dissimilarity of distributions to some reference distributions denoted as templates. Our framework exte…

Inferring the structural properties of a protein from its amino acid sequence is a challenging yet important problem in biology. Structures are not known for the vast majority of protein sequences, but structure is critical for understanding function. Existing approaches for detecting structural similarity between prot…

We consider the problem of optimal recovery of true ranking of $n$ items from a randomly chosen subset of their pairwise preferences. It is well known that without any further assumption, one requires a sample size of $Ω(n^2)$ for the purpose. We analyze the problem with an additional structure of relational graph $G([…

There is a growing need for discrete choice models that account for the complex nature of human choices, escaping traditional behavioral assumptions such as the transitivity of pairwise preferences. Recently, several parametric models of intransitive comparisons have been proposed, but in all cases the maximum likeliho…

Transformers learn topic structure through embedding and attention mechanisms.

The paper studies 3-manifolds with specific Morse-Smale diffeomorphisms and finds they are homeomorphic to lens spaces.

The study develops a word mechanism for knot and link diagrams.

This paper examines the problem of ranking a collection of objects using pairwise comparisons (rankings of two objects). In general, the ranking of $n$ objects can be identified by standard sorting methods using $n log_2 n$ pairwise comparisons. We are interested in natural situations in which relationships among the o…

Vector embedding is a foundational building block of many deep learning models, especially in natural language processing. In this paper, we present a theoretical framework for understanding the effect of dimensionality on vector embeddings. We observe that the distributional hypothesis, a governing principle of statis…

p-SNE embeds Poisson count data into low dimensions preserving structure.

Embedding-based Knowledge Base Completion models have so far mostly combined distributed representations of individual entities or relations to compute truth scores of missing links. Facts can however also be represented using pairwise embeddings, i.e. embeddings for pairs of entities and relations. In this paper we ex…

New symplectic barriers found in ball embeddings.

Unified framework for SSL methods linking contrastive and non-contrastive approaches.

Study on pairwise counter-monotonicity, a type of negative dependence.

Study shows attention-style models learn pairwise interactions efficiently.

Given a simply-connected closed 4-manifold $X$ and a smoothly embedded oriented surface $Σ$, various constructions based on Fintushel-Stern knot surgery have produced new surfaces in $X$ that are pairwise homeomorphic to $Σ$, but not diffeomorphic. We prove that for all known examples of surface knots constructed from …

The study proves exotic smooth structures and equivalent genus functions in 4-manifolds.

We propose a novel node embedding of directed graphs to statistical manifolds, which is based on a global minimization of pairwise relative entropy and graph geodesics in a non-linear way. Each node is encoded with a probability density function over a measurable space. Furthermore, we analyze the connection between th…

This paper proposes a variant of the method of Guédon and Verhynin for estimating the cluster matrix in the Mixture of Gaussians framework via Semi-Definite Programming. A clustering oriented embedding is deduced from this estimate. The procedure is suitable for very high dimensional data because it is based on pairwis…

PGRec improves recommendation by modeling user-item preferences as a graph and embedding it for better predictions.

We present an infinite sequence of smooth embeddings of a connected sum of 6 projective planes in the 4-sphere, which are all ambient homeomorphic, but pairwise ambient non-diffeomorphic. The double covers of the 4-sphere ramified along these surfaces form a family of the exotic $\Bbb CP^2#5\bar{\Bbb CP^2}$ constructed…

Automatic cover detection -- the task of finding in an audio database all the covers of one or several query tracks -- has long been seen as a challenging theoretical problem in the MIR community and as an acute practical problem for authors and composers societies. Original algorithms proposed for this task have prove…

In this paper, we provide a theoretical understanding of word embedding and its dimensionality. Motivated by the unitary-invariance of word embedding, we propose the Pairwise Inner Product (PIP) loss, a novel metric on the dissimilarity between word embeddings. Using techniques from matrix perturbation theory, we revea…

New method detects communities in hypergraphs by embedding them into a vector space.

We propose a hierarchical correlation clustering method that extends the well-known correlation clustering to produce hierarchical clusters applicable to both positive and negative pairwise dissimilarities. Then, in the following, we study unsupervised representation learning with such hierarchical correlation clusteri…

Representation learning on networks offers a powerful alternative to the oft painstaking process of manual feature engineering, and as a result, has enjoyed considerable success in recent years. However, all the existing representation learning methods are based on the first-order network (FON), that is, the network th…

GNNRank uses neural networks to learn global rankings from competition match data.

Study shows how to create special metrics on 4-manifolds with certain spheres.

A new method for name disambiguation in academic networks using multi-view attention and recurrent neural networks.

For a topological space $X$ we study continuous maps $f : X\to \mathbb R^m$ such that images of every pairwise distinct $k$ points are affinely (linearly) independent. Such maps are called affinely (linearly) $k$-regular embeddings. We investigate the cohomology obstructions to existence of regular embeddings and give …