Optimal kernel sum classifiers analyzed for statistical efficiency.
problem Analyzing the statistical efficiency of optimal kernel sum classifiers.
method Combining optimization tools with learning theory bounds to analyze sample complexity.
result Justifies assumptions in prior work on multiple kernel learning and provides a new form of Rademacher complexity.
Similarity-based clustering and semi-supervised learning methods separate the data into clusters or classes according to the pairwise similarity between the data, and the pairwise similarity is crucial for their performance. In this paper, we propose a novel discriminative similarity learning framework which learns dis…
In this paper, we are interested in constructing general graph-based regularizers for multiple kernel learning (MKL) given a structure which is used to describe the way of combining basis kernels. Such structures are represented by sum-product networks (SPNs) in our method. Accordingly we propose a new convex regulariz…
New kernels on symmetric groups enable efficient Gaussian process sampling.
problem Efficiently modeling and sampling on symmetric groups.
method Introduced power sum kernels and methods for efficient calculation and sampling.
result Polynomial computational complexity for sampling Gaussian processes.
Study integral kernels on complex symmetric spaces and their Dyson Brownian Motion applications.
problem Analysis of integral kernels on complex symmetric spaces.
method Simple new method of alternating sum formulas to construct W-invariant kernels and their asymptotic behavior. result Obtained asymptotic behavior of integral kernels and applied to Dyson Brownian Motion.
KSOS improves kernel learning for dynamical systems via global optimization.
problem Challenges in selecting optimal kernels and tuning parameters in traditional kernel-based methods.
method Global optimization framework with kernel-based surrogate functions.
result KSOS consistently outperforms gradient descent in predicting dynamical systems.
This note optimizes distributions using kernel mean embeddings with a new parameterization.
problem Optimizing distributions using kernel mean embeddings is challenging due to the difficulty of characterizing probability distribution vectors.
method Proposes a new parameterization of positive functions using kernel sums-of-squares to fit distributions in the MMD geometry.
result Distributions with kernel sum-of-squares densities are dense in the MMD geometry, allowing optimization in the finite-sample setting.
We provide a fast approximation to eNTKs for neural networks.
problem Efficiently computing eNTKs for large networks.
method Developed and proved the 'sum of logits' approximation.
result The 'sum of logits' approximation converges to eNTKs at initialization.
New classifier combines locally linear kernels for fast and accurate non-linear classification.
problem Developing a fast and accurate non-linear classifier.
method Combines locally linear classifiers using a ℓ1 Multiple Kernel Learning (MKL) problem with scalable MKL training for streaming kernels. result The resulting classifier achieves high accuracy with fast inference time.
Paper introduces a new kernel model for PSD-valued functions with theoretical guarantees and applications.
problem Enforcing positive semi-definiteness (PSD) in function models with good performance and theoretical guarantees.
method Kernel sum-of-squares model for PSD-valued functions, extending previous models for non-negative scalar functions.
result The model constitutes a universal approximator of PSD functions and can represent any smooth and strongly convex function.
A new method slices and sums radial kernels faster.
problem Fast computation of large kernel sums in kernel methods.
method Random projections to 1D subspaces and QMC for selecting projections.
result QMC-slicing outperforms existing methods on test datasets.
The paper analyzes kernel classifiers' performance in Sobolev spaces and proves their optimality.
problem Theoretical analysis of kernel classifiers' performance in Sobolev spaces.
method Deriving upper and lower bounds on classification excess risk using kernel regression theory and estimating interpolation smoothness.
result The proposed kernel classifier is optimal in Sobolev spaces, with theoretical bounds confirmed by real data.
Novel Newton method for large-scale kernel methods using random features.
problem Efficiently solving large-scale finite-sum minimization problems in RKHS.
method Randomized feature-based Newton method for empirical risk minimization.
result Local superlinear and global linear convergence of the method.
This paper presents a novel kernel-based generative classifier which is defined in a distortion subspace using polynomial series expansion, named Kernel-Distortion (KD) classifier. An iterative kernel selection algorithm is developed to steadily improve classification performance by repeatedly removing and adding kerne…
Quantum models use complex Hilbert spaces for uncertainty.
problem Modeling dynamics in continuous-valued features.
method Quantum Graphical Models (QGMs) and Hilbert Space Embedding (HSE).
result HSE-HQMMs are competitive with state-of-the-art models.
Defines a new algebra for singular foliations, extending Schwartz kernels.
problem Extending Schwartz kernel operators to singular foliations.
method Defines convolution algebra of transverse distributions, proves representation as operators on spaces of functions.
result Generalizes Schwartz kernel operators to singular foliations.
A typical approach in estimating the learning rate of a regularized learning scheme is to bound the approximation error by the sum of the sampling error, the hypothesis error and the regularization error. Using a reproducing kernel space that satisfies the linear representer theorem brings the advantage of discarding t…
Kernel Bayesian inference is a principled approach to nonparametric inference in probabilistic graphical models, where probabilistic relationships between variables are learned from data in a nonparametric manner. Various algorithms of kernel Bayesian inference have been developed by combining kernelized basic probabil…
The medical research facilitates to acquire a diverse type of data from the same individual for particular cancer. Recent studies show that utilizing such diverse data results in more accurate predictions. The major challenge faced is how to utilize such diverse data sets in an effective way. In this paper, we introduc…
Improved KELM for multiclass classification with wavelet kernel.
problem Low test accuracy in multiclass classification problems.
method Mexican Hat wavelet kernel ELM.
result Significantly improved performance compared to other classifiers.
Formula for Bergman kernel of complex hyperbolic manifolds proved.
problem Calculating the Bergman kernel of complex hyperbolic manifolds.
method Expressed as a sum over geodesic loops.
result Maximum and minimum of the Bergman kernel function proved.
We investigate iterated compositions of weighted sums of Gaussian kernels and provide an interpretation of the construction that shows some similarities with the architectures of deep neural networks. On the theoretical side, we show that these kernels are universal and that SVMs using these kernels are universally con…
The extended Wild sums considered in this article generalize the classi- cal Wild sums of statistical physics. We first show how to obtain explicit solutions for the evolution equation of a large system where the interactions are given by a single, but general, interacting kernel which involves m components, for a fixe…
Support vector data description (SVDD) is a popular technique for detecting anomalies. The SVDD classifier partitions the whole space into an inlier region, which consists of the region near the training data, and an outlier region, which consists of points away from the training data. The computation of the SVDD class…
Inspired by a growing interest in analyzing network data, we study the problem of node classification on graphs, focusing on approaches based on kernel machines. Conventionally, kernel machines are linear classifiers in the implicit feature space. We argue that linear classification in the feature space of kernels comm…
A new method extracts features from time series data using iterated sums and improves classification accuracy.
problem Time series classification challenges.
method Feature extraction using iterated-sums signature (ISS) followed by a linear classifier.
result Competitive with state-of-the-art methods on UCR archive.
Classifies exceptional Legendrian realizations of Hopf link connected sums.
problem Classifying exceptional Legendrian realizations of Hopf link connected sums.
method Complete coarse classification using Legendrian knot theory.
result First classification result about exceptional Legendrian representatives for Hopf link connected sums.
Paper introduces MinDiff framework for balancing classifier performance and fairness.
problem Balancing classifier performance and fairness in machine learning models.
method MinDiff framework with kernel-based statistical dependency tests.
result Demonstrates real-world improvements in classifier performance and fairness.
Proposes a method to improve transparency and robustness of deep learning models.
problem Lack of interpretability and robustness in deep learning models.
method A weighted sum of training instances with learned instance-embedding space weights.
result Improved transparency, controlled error rates, and robustness to out-of-domain data.
New string kernels discover global properties through random feature maps, avoiding quadratic complexity.
problem Existing string kernels struggle with capturing long patterns, maintaining positive definiteness, and handling large datasets efficiently.
method Proposes a new class of global string kernels using random feature maps to discover global properties through global alignments, ensuring positive definiteness and linear computational cost.
result Random String Embeddings (RSE) achieve better or comparable accuracy to state-of-the-art methods, especially for longer strings.
PAC-Bayesian bounds improve understanding of K-NN classifier performance.
problem Improving the understanding of K-NN classifier's generalization error.
method PAC-Bayesian analysis applied to K-NN classifier in kernel space.
result PAC-Bayesian bounds provide a function of the number of redundant training examples.
Study provides bounds for estimating intrinsic dimension using Gaussian kernels.
problem Estimating intrinsic dimension from data.
method Finite-sample concentration and anti-concentration bounds for Gaussian kernel sums.
result Explicit dependence on sample size, bandwidth, and geometric parameters.
KOC+ uses privileged information to improve one-class classification performance.
problem Outlier detection and novelty detection using kernel methods.
method Kernel ridge regression with correction function for privileged information.
result KOC+ achieves better generalization performance compared to traditional methods.
Optimal scoring framework for kernel classification with feature selection.
problem Two-group classification problem.
method Optimal scoring framework, structured sparsity using weighted kernels, automated parameter selection.
result Superior classification performance compared to existing nonparametric classifiers.
Generalizes PCA to maximize any convex function of components.
problem Finding a principal vector that maximizes a convex function of components.
method Gradient ascent algorithm for solving the generalized PCA problem; fixed points of neural networks for kernel version.
result Solutions can be obtained as fixed points of simple neural networks.
New kernel method for shape classification on Kendall shape space.
problem Classification of shapes on non-Euclidean Kendall shape space.
method Extrinsic Veronese Whitney Gaussian kernel for KRRC on Σ2k. result KRRC classifier performs well on real Kendall shape data.
Paper studies multiclass classifiers from binary classifiers, proving methods and demonstrating advantages.
problem Constructing efficient multiclass classifiers from binary ones.
method Two methods: one vs. all and hierarchical classification, with a new leverage-hierarchical method introduced.
result Proves upper bounds and exact formulas for multiclass regret in terms of binary regrets.
The paper classifies 3D self-expanders with specific properties.
problem Classifying self-expanders with constant curvature components.
method Complete classification of 3D self-expanders with specific curvature conditions.
result Completely classified 3D self-expanders with constant curvature components.
Enhances classifier performance through feature space transformations and model selection.
problem Improving the accuracy of classifiers by reducing complexity.
method Combining feature mapping, prototype selection, and kernel function transformations to transform data into a more convenient distribution.
result Our methods produce competitive classifiers and are statistically different among them.
Large filters improve performance but are costly; this work uses learned box filters and summed-area tables.
problem Improving performance in dense prediction tasks like human pose estimation with large filters.
method Adopted learnable box filters and summed-area tables to reduce computational cost and maintain performance.
result Demonstrated competitive performance on human pose estimation benchmarks.
New method selects kernel bandwidth for SVDD and OCSVM.
problem Selecting optimal Gaussian kernel bandwidth for SVDD and OCSVM.
method Exploits low-rank representation of kernel matrix to suggest bandwidth.
result Method performs well for both low-dimensional and high-dimensional data.
We study Legendrian singular links up to contact isotopy. Using a special property of the singular points, we define the singular connected sum of Legendrian singular links. This concept is a generalization of the connected sum and can be interpreted as a tangle replacement, which provides a way to classify Legendrian …
Support vector machines (SVM) and other kernel techniques represent a family of powerful statistical classification methods with high accuracy and broad applicability. Because they use all or a significant portion of the training data, however, they can be slow, especially for large problems. Piecewise linear classifie…
New method uses path signatures for efficient likelihood estimation in time-series data.
problem Intractable likelihood functions in complex dynamic models.
method Kernel classifier based on path signatures for sequential data.
result Path signatures yield highly performant classifiers, even with low sample numbers.
This study examines when non-parametric methods are robust to adversarial examples.
problem Understanding when non-parametric methods are robust to adversarial examples.
method Examined general non-parametric methods and established conditions for r-consistency.
result Non-parametric methods like nearest neighbors and kernel classifiers are r-consistent when data is well-separated, while histograms are not.
We study the behavior of Legendrian and transverse knots under the operation of connected sums. As a consequence we show that there exist Legendrian knots that are not distinguished by any known invariant. Moreover, we classify Legendrian knots in some non-Legendrian simple knot types.
New method finds global minima using function evaluations and kernel approximations.
problem Finding global minima of smooth functions with limited evaluations.
method Approximates the function using infinite sums of square smooth functions and solves the optimization problem with polynomial time complexity.
result Achieves optimal number of function evaluations with theoretical guarantees and nearly optimal convergence rate.
We show that the Artin representation on concordance classes of string links induces a well-defined epimorphism modulo order n twisted Whitney tower concordance, and that the kernel of this map is generated by band sums of iterated Bing-doubles of any string knot with nonzero Arf invariant. We also continue J. Levine's…