arXiv research

EDAs with matrix transpose improve Bayesian structure learning performance.

Missing data estimation is an important challenge with high-dimensional data arranged in the form of a matrix. Typically this data matrix is transposable, meaning that either the rows, columns or both can be treated as features. To model transposable data, we present a modification of the matrix-variate normal, the mea…

We consider the problem of large-scale inference on the row or column variables of data in the form of a matrix. Often this data is transposable, meaning that both the row variables and column variables are of potential interest. An example of this scenario is detecting significant genes in microarrays when the samples…

A new method for higher-order co-occurrences in hypergraphs.

Transposable data represents interactions among two sets of entities, and are typically represented as a matrix containing the known interaction values. Additional side information may consist of feature vectors specific to entities corresponding to the rows and/or columns of such a matrix. Further information may also…

Graph embedding learns low-dimensional representations for nodes in a graph and effectively preserves the graph structure. Recently, a significant amount of progress has been made toward this emerging research area. However, there are several fundamental problems that remain open. First, existing methods fail to preser…

The paper introduces transposed Poisson superalgebras and their properties.

Given a knot and an SL(n,C) representation of its group that is conjugate to its dual, the representation that replaces each matrix with its inverse-transpose, the associated twisted Reidemeister torsion is reciprocal. An example is given of a knot group and SL(3,Z) representation that is not conjugate to its dual for …

New Sasaki-Einstein 7-spheres found via Berglund-Hübsch transpose.

Introduces $q$-transpose for $q$-deformed modular group matrices.

Combines gradient-based and competitive learning for unsupervised feature extraction.

We introduce a guide to help deep learning practitioners understand and manipulate convolutional neural network architectures. The guide clarifies the relationship between various properties (input shape, kernel shape, zero padding, strides and output shape) of convolutional, pooling and transposed convolutional layers…

Kernel clustering algorithm improved for large datasets using incomplete Cholesky factorization.

Study explores K-means clustering of variables and its relation to PCA.

The Berglund-Hübsch rule connects Calabi-Yau orbifolds to Sasakian manifolds.

Variables in many massive high-dimensional data sets are structured, arising for example from measurements on a regular grid as in imaging and time series or from spatial-temporal measurements as in climate studies. Classical multivariate techniques ignore these structural relationships often resulting in poor performa…

This paper studies iteration convergence of Kronecker graphical lasso (KGLasso) algorithms for estimating the covariance of an i.i.d. Gaussian random sample under a sparse Kronecker-product covariance model and MSE convergence rates. The KGlasso model, originally called the transposable regularized covariance model by …

New algorithm estimates robust Gaussian covariance in nearly matrix multiplication time.

Language models fail to process hallucinated responses, and this study diagnoses the failure.

We study the Low Rank Phase Retrieval (LRPR) problem defined as follows: recover an $n \times q$ matrix $X^*$ of rank $r$ from a different and independent set of $m$ phaseless (magnitude-only) linear projections of each of its columns. To be precise, we need to recover $X^*$ from $y_k := |A_k{}' x^*_k|, k=1,2,\dots, q$…

New examples show Sasaki manifolds without extremal metrics.

Tensor programs prove neural network limits for any architecture.

Transposable Elements (TEs) or jumping genes are the DNA sequences that have an intrinsic capability to move within a host genome from one genomic location to another. Studies show that the presence of a TE within or adjacent to a functional gene may alter its expression. TEs can also cause an increase in the rate of m…

This work bridges competitive learning with gradient-based learning for faster feature extraction.

Let G be a connected bipartite graph with color classes E and V and root polytope Q. Regarding the hypergraph (V,E) induced by G, we prove that its interior polynomial is equivalent to the Ehrhart polynomial of Q, which in turn is equivalent to the h-vector of any triangulation of Q. It follows that the interior polyno…

We propose a semantic segmentation model that exploits rotation and reflection symmetries. We demonstrate significant gains in sample efficiency due to increased weight sharing, as well as improvements in robustness to symmetry transformations. The group equivariant CNN framework is extended for segmentation by introdu…

We propose a deep factorization model for typographic analysis that disentangles content from style. Specifically, a variational inference procedure factors each training glyph into the combination of a character-specific content embedding and a latent font-specific style variable. The underlying generative model combi…

Most existing GANs architectures that generate images use transposed convolution or resize-convolution as their upsampling algorithm from lower to higher resolution feature maps in the generator. We argue that this kind of fixed operation is problematic for GANs to model objects that have very different visual appearan…

We study Nijenhuis structures on Courant algebroids in terms of the canonical Poisson bracket on their symplectic realizations. We prove that the Nijenhuis torsion of a skew-symmetric endomorphism N of a Courant algebroid is skew-symmetric if the square of N is proportional to the identity, and only in this case when t…

The paper extends Gray's result to quaternion-Kähler manifolds.

A new and an enriched JPEG algorithm is provided for identifying redundancies in a sequence of irregular noisy data points which also accommodates a reference-free criterion function. Our main contribution is by formulating analytically (instead of approximating) the inverse of the transpose of JPEGwavelet transform wi…

A hybrid ASR system using conformer architecture improves word-error-rate and training speed.

New Sliced-Wasserstein distances for non-Euclidean data.

Gatherings of thousands to millions of people frequently occur for an enormous variety of events, and automated counting of these high-density crowds is useful for safety, management, and measuring significance of an event. In this work, we show that the regularly accepted labeling scheme of crowd density maps for trai…

In classical General Relativity, the way to exhibit the equations for the gravitational waves is based on two "tricks" allowing to transform the Einstein equations after linearizing them over the Minkowski metric. With specific notations used in the study of {\it Lie pseudogroups} of transformations of an $n$-dimension…

Study on feature learning dynamics in infinite-depth neural networks, focusing on ResNets.

Orthogonium offers unified, efficient layers for robust deep learning.

The variational autoencoder (VAE) is a popular probabilistic generative model. However, one shortcoming of VAEs is that the latent variables cannot be discrete, which makes it difficult to generate data from different modes of a distribution. Here, we propose an extension of the VAE framework that incorporates a classi…

This paper extends the evolution operator to contact mechanics, linking Lagrangian and Hamiltonian formulations.

RFSVM uses learned RF similarity for HDLSS classification.

Proposes a fixed smooth convolutional layer to reduce checkerboard artifacts in CNNs.

To date, attribute discretization is typically performed by replacing the original set of continuous features with a transposed set of discrete ones. This paper provides support for a new idea that discretized features should often be used in addition to existing features and as such, datasets should be extended, and n…

The concept of explainability is envisioned to satisfy society's demands for transparency on machine learning decisions. The concept is simple: like humans, algorithms should explain the rationale behind their decisions so that their fairness can be assessed. While this approach is promising in a local context (e.g. to…

Neural networks are commonly trained to make predictions through learning algorithms. Contrastive Hebbian learning, which is a powerful rule inspired by gradient backpropagation, is based on Hebb's rule and the contrastive divergence algorithm. It operates in two phases, the forward (or free) phase, where the data are …

Study local moduli of Sasaki-Einstein metrics on specific polynomial links.

Proposes continuous convolution layers for flexible feature map resizing.

Efficiently explains model outputs using HSIC, a dependence measure.

Training state-of-the-art, deep neural networks is computationally expensive. One way to reduce the training time is to normalize the activities of the neurons. A recently introduced technique called batch normalization uses the distribution of the summed input to a neuron over a mini-batch of training cases to compute…