Deep neural networks for structured prediction using kernel-induced losses.
problem Structured prediction tasks for images and texts.
method Designing a novel family of deep neural architectures that predict in a finite-dimensional subspace derived from the kernel-induced loss.
result Gradient descent algorithms can be used for structured prediction with deep neural networks.
DGKIP extends KIP for dataset distillation without bi-level optimization.
problem Efficiently distill datasets for various loss functions.
method Leverages duality theory to avoid bi-level optimization.
result DGKIP supports a wider range of loss functions.
Kernel Induced Random Survival Forests (KIRSF) is a statistical learning algorithm which aims to improve prediction accuracy for survival data. As in Random Survival Forests (RSF), Cumulative Hazard Function is predicted for each individual in the test set. Prediction error is estimated using Harrell's concordance inde…
A new method for privacy-preserving data distillation using wavelet features from ScatterNet.
problem Privacy-preserving data distillation for small datasets.
method Differentially private kernel inducing points using wavelet features from ScatterNet.
result Achieves privacy-preserving data distillation with manageable computational cost.
SkyGP improves Gaussian process scalability for real-time learning.
problem Scalability issues with exact Gaussian processes for streaming data.
method Streaming kernel-induced progressively generated Gaussian process experts (SkyGP).
result SkyGP maintains performance guarantees while improving scalability.
A new method compresses conditional distributions of labelled data.
problem No existing method directly compresses the conditional distribution of labelled data.
method Introduce Average Maximum Conditional Mean Discrepancy (AMCMD), derive a closed form estimator, and extend Kernel Herding (KH) to Average Conditional Kernel Herding (ACKH).
result Directly compressing conditional distributions outperforms joint distribution compression and greedy selection.
Stochastic kernel based dimensionality reduction approaches have become popular in the last decade. The central component of many of these methods is a symmetric kernel that quantifies the vicinity between pairs of data points and a kernel-induced Markov chain on the data. Typically, the Markov chain is fully specified…
Deep Gaussian Processes with polynomial kernels can collapse rapidly without proper hyperparameter tuning.
problem The collapse of Deep Gaussian Processes with polynomial kernels without careful hyperparameter tuning.
method Analysis using the Berry-Esseen Theorem and observation of prior behavior.
result The prior of a Deep Gaussian Process collapses rapidly towards zero or places negligible mass on low norm functions without proper hyperparameter tuning.
Proposes DAK model for improved GP computations.
problem Challenges in high-dimensional GP layers in DKL.
method Additive structure and induced prior approximation for GP units.
result Outperforms state-of-the-art DKL methods in regression and classification.
Proposes a TS approach for Bayesian optimization with preferential feedback.
problem Optimizing with preference feedback in complex applications.
method Uses Thompson Sampling with a dueling kernel and anchor invariance.
result Performance matches standard TS for scalar feedback in finite time.
New kernels capture both local and non-local interactions efficiently.
problem Designing kernels that capture both local and non-local interactions while remaining computationally tractable.
method Spectral truncation kernels based on C∗-algebra. result Spectral truncation kernels induce interactions across the data function domain and reduce computational cost.
We provide a unifying framework linking two classes of statistics used in two-sample and independence testing: on the one hand, the energy distances and distance covariances from the statistics literature; on the other, distances between embeddings of distributions to reproducing kernel Hilbert spaces (RKHS), as establ…
Proposes a framework to extract ordered eigenfunctions from contextual kernels.
problem Lack of exact spectral decomposition in existing methods.
method Modular building blocks for compatibility with contextual kernels and scalability.
result Extracted eigenfunctions provide effective importance scores for feature selection.
KIP meta-learning compresses datasets significantly.
problem Training data size and quality issues in machine learning.
method Kernel Inducing Points (KIP) for dataset compression.
result Significant reduction in dataset size with similar model performance.
GD-trained shallow ReLU nets learn Lipschitz functions with noise.
problem Learning Lipschitz functions with additive noise in overparameterized neural networks.
method Gradient Descent (GD) with early stopping, focusing on the Neural Tangent Kernel (NTK).
result Early-stopped GD achieves minimax optimal rates for learning Lipschitz functions.
Two-layer neural networks learn efficiently using kernel methods in mean-field analysis.
problem Feature learning ability of two-layer neural networks in the mean-field regime.
method Mean-field analysis through kernel methods, focusing on dynamics of the first layer's kernel.
result Two-layer neural networks can learn a union of multiple reproducing kernel Hilbert spaces more efficiently than kernel methods.
KMRCD detects outliers in non-elliptical data using kernel trick.
problem Outlier detection in non-elliptical data.
method KMRCD estimator that uses kernel trick to compute robust covariance matrix in a feature space.
result KMRCD performs well in simulations and real-life data.
DKLM learns adaptive kernels for robust nonlinear subspace clustering.
problem Nonlinear structures in data and challenges with kernel-based clustering.
method Data-driven kernel learning with adaptive weighting and optimal block-diagonal affinity matrix.
result DKLM enhances robustness and preserves manifold structure in nonlinear space.
RFAD uses random features to speed up dataset distillation.
problem Efficiently compress large datasets for reduced storage and computation.
method Random feature approximation of the Neural Network Gaussian Process kernel.
result At least 100-fold speedup over KIP with competitive accuracy.
We propose a novel hierarchical model for multitask bipartite ranking. The proposed approach combines a matrix-variate Gaussian process with a generative model for task-wise bipartite ranking. In addition, we employ a novel trace constrained variational inference approach to impose low rank structure on the posterior m…
Bayesian neural networks fail at OOD detection, revealing fundamental issues.
problem The ability of Bayesian neural networks to detect out-of-distribution data.
method Empirical study of Bayesian inference in neural networks, including infinite-width and finite-width cases using Hamiltonian Monte Carlo.
result Bayesian inference in common neural network architectures does not lead to good OOD detection.
A new framework learns differentiable structured losses from data.
problem Learning effective losses for complex structured prediction tasks.
method Contrastive learning to learn differentiable structured losses from output data.
result Achieves similar or better performance than kernel-based methods.
Decision forests are widely used for classification and regression tasks. A lesser known property of tree-based methods is that one can construct a proximity matrix from the tree(s), and these proximity matrices are induced kernels. While there has been extensive research on the applications and properties of kernels, …
The diversification (generating slightly varying separating discriminators) of Support Vector Machines (SVMs) for boosting has proven to be a challenge due to the strong learning nature of SVMs. Based on the insight that perturbing the SVM kernel may help in diversifying SVMs, we propose two kernel perturbation based b…
A new method for distribution regression using sliced Wasserstein distance.
problem Learning functions over spaces of probabilities.
method Proposes an OT-based estimator using the Sliced Wasserstein distance.
result Proves universal consistency and excess risk bounds for the proposed estimator.
New methods for ordinal classification of interval-valued data and functional data.
problem Ordinal classification of interval-valued data and functional data.
method Six ordinal classifiers are proposed, including parametric, binary decomposition, logistic regression, distance-based, k-nearest-neighbor, kernel PCA, and random forest methods.
result Considering ordering and interval-valued information improves the accuracy of ordinal classification.
Uniform bounds for neural networks' generalization error in overparameterized settings.
problem Generalization error in overparameterized neural networks.
method Neural Tangent kernel theory and Mercer decomposition of the NT kernel in spherical harmonics.
result Uniform generalization bounds for overparameterized neural networks in RKHS.
Geometric regularisation improves statistical models by avoiding degeneracy loci.
problem Non-identifiability, singular information, and moment indeterminacy in statistical models.
method Develops the geometric regularisation of distribution-kernel pairs (T,φ) using Whitney, Thom, and Mather theorems. result Finite-dimensional weak transversality theorem for generic kernels, avoiding degeneracy strata of high codimension.
A new algorithm reduces sample complexity for learning Q-functions in reinforcement learning.
problem Efficiently learning Q-functions in reinforcement learning with continuous state and action spaces.
method Developed a simple, iterative learning algorithm that estimates low-rank Q-functions.
result Achieved exponential improvement in sample complexity for low-rank Q-functions.
This work challenges the Neural Tangent Kernel's role in overparameterized neural networks, especially with large width and depth.
problem The Neural Tangent Kernel's behavior in overparameterized neural networks with large width and depth is unclear.
method Experimental and theoretical analysis of ReLU networks with large width and depth.
result The aggregate norm of hidden neuron deviations does not vanish in infinitely-wide ReLU networks, indicating non-trivial behavior.