arXiv research

This paper introduces new loss functions for balanced multi-class classification.

The paper proves neural networks' consistency and optimal convergence rates for various function classes.

We study consistency of learning algorithms for a multi-class performance metric that is a non-decomposable function of the confusion matrix of a classifier and cannot be expressed as a sum of losses on individual data points; examples of such performance metrics include the macro F-measure popular in information retri…

New method learns multi-class from single-class data with confidences.

Study on top-$k$ classification with new loss functions and algorithms.

We introduce Fisher consistency in the sense of unbiasedness as a desirable property for estimators of class prior probabilities. Lack of Fisher consistency could be used as a criterion to dismiss estimators that are unlikely to deliver precise estimates in test datasets under prior probability and more general dataset…

New algorithm estimates causal effects for non-Gaussian data.

We investigate relation between Dehn fillings and commensurability of hyperbolic 3-manifolds. The set consisting of the commensurability classes of hyperbolic 3-manifolds admits the quotient topology induced by the geometric topology. We show that this quotient space satisfies some separation axioms. Roughly speaking, …

OwMatch improves open-world semi-supervised learning by self-labeling and consistency.

Enhanced $H$-consistency bounds derived under relaxed conditions.

Study on $H$-consistency bounds for machine learning surrogates.

Improves GAN-based semi-supervised learning with consistency regularization.

Study dynamic risk measures and performance indices using distortion functions.

In many countries information on expectations collected through consumer confidence surveys are used in macroeconomic policy formulation. Unfortunately, before doing so, the consistency of responses is often not taken into account, leading to biases creeping in and affecting the reliability of the indices hence created…

New method interpolates training data and is consistent for various data distributions.

We consider the problem of $n$-class classification ($n\geq 2$), where the classifier can choose to abstain from making predictions at a given cost, say, a factor $α$ of the cost of misclassification. Designing consistent algorithms for such $n$-class classification problems with a `reject option' is the main goal of t…

In this work we show that, using the eigen-decomposition of the adjacency matrix, we can consistently estimate feature maps for latent position graphs with positive definite link function $κ$, provided that the latent positions are i.i.d. from some distribution F. We then consider the exploitation task of vertex classi…

Reduces bounded loss learning to binary classification.

Simplifies online learning with consistent oracle to fewer mistakes.

New algorithms for regression with adversarial responses on various metric spaces.

Causal graphs, such as directed acyclic graphs (DAGs) and partial ancestral graphs (PAGs), represent causal relationships among variables in a model. Methods exist for learning DAGs and PAGs from data and for converting DAGs to PAGs. However, these methods are significantly limited in that they only output a single cau…

Generates spin structure stabilizers using Dehn twists.

Deep Convolutional Neural Networks (CNN) enforces supervised information only at the output layer, and hidden layers are trained by back propagating the prediction error from the output layer without explicit supervision. We propose a supervised feature learning approach, Label Consistent Neural Network, which enforces…

Paper tackles class-imbalanced semi-supervised learning, improving performance.

Study on tradeoffs between mistakes and ERM oracle calls in online and transductive learning.

We address a class of schemes for the Euler equations with the following features: the space discretization is staggered, possible upwinding is performed with respect to the material velocity only and the internal energy balance is solved, with a correction term designed on consistency arguments. These schemes have bee…

New algorithms for latent class analysis using regularized spectral clustering.

Gini index needs auto-calibration for consistent decision-making.

This thesis tackles learning with multi-class abstention and multi-expert deferral, improving model reliability and efficiency.

We obtain a finite set of generators for the level 2 mapping class group of a closed nonorientable surface of genus $g\ge 3$. This set consists of isotopy classes of Lickorish's Y-homeomorphisms also called crosscap slides.

New method simplifies checking consistency of differentiable loss functions.

Transferring learned models to novel tasks is a challenging problem, particularly if only very few labeled examples are available. Although this few-shot learning setup has received a lot of attention recently, most proposed methods focus on discriminating novel classes only. Instead, we consider the extended setup of …

New algorithm achieves consistent learning from context in bandit problems.

Paper connects risk consistency to L_p consistency for broader loss functions.

We give an infinite presentation for the mapping class group of a non-orientable surface. The generating set consists of all Dehn twists and all crosscap pushing maps along simple loops.

Triangular flows ensure statistical consistency and fast rates in generative modeling.

Variable screening is a fast dimension reduction technique for assisting high dimensional feature selection. As a preselection method, it selects a moderate size subset of candidate variables for further refining via feature selection to produce the final model. The performance of variable screening depends on both com…

Proof of Gaussian ML estimator consistency in linear auto-regressive models.

We characterize the class of exchangeable feature allocations assigning probability $V_{n,k}\prod_{l=1}^{k}W_{m_{l}}U_{n-m_{l}}$ to a feature allocation of $n$ individuals, displaying $k$ features with counts $(m_{1},\ldots,m_{k})$ for these features. Each element of this class is parametrized by a countable matrix $V$…

For G_2-manifolds the Fernández-Gray class X_1+X_4 is shown to consist of the union of the class X_4 of G_2-manifolds locally conformal to parallel G_2-structures and that of conformal transformations of nearly parallel or weak holonomy G_2-manifolds of type X_1. The analogous conclusion is obtained for Gray-Hervella c…

In this paper we study the consistency of an empirical minimum error entropy (MEE) algorithm in a regression setting. We introduce two types of consistency. The error entropy consistency, which requires the error entropy of the learned function to approximate the minimum error entropy, is shown to be always true if the…

The Nearest subspace classifier (NSS) finds an estimation of the underlying subspace within each class and assigns data points to the class that corresponds to its nearest subspace. This paper mainly studies how well NSS can be generalized to new samples. It is proved that NSS is strongly consistent under certain assum…

This survey consists of a detailed proof of Markov's Theorem based on Joan Birman's book "Braids, Links, and Mapping Class Groups" and Carlo Petronio's classes. It was part of an exam project in A.Y. 2016/2017 for the course Knot Theory.

New method proves neural networks can select features consistently.

We propose a new class of mappings, called Dynamic Limit Growth Indices, that are designed to measure the long-run performance of a financial portfolio in discrete time setup. We study various important properties for this new class of measures, and in particular, we provide necessary and sufficient condition for a Dyn…

Bandit algorithms struggle with consistent performance and robustness.

This paper rethinks confidence calibration under covariate shifts.

New framework infers multiple classes per image for one-shot learning.