Knowledge distillation is effective for producing small, high-performance neural networks for classification, but these small networks are vulnerable to adversarial attacks. This paper studies how adversarial robustness transfers from teacher to student during knowledge distillation. We find that a large amount of robu…
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
168,657 papers · 148 categories
Trend · papers per month
48 results for “ARD network”
In this work, we utilize Machine Learning for early recognition of patients at high risk of acute respiratory distress syndrome (ARDS), which is critical for successful prevention strategies for this devastating syndrome. The difficulty in early ARDS recognition stems from its complex and heterogenous nature. In this s…
Deep state space model forecasts time series with uncertainty.
problem Probabilistic forecasting for risk management.
method Parameterized deep networks for non-linear models, recurrent neural nets for dependency, ARD network for exogenous variables.
result Accurate and sharp probabilistic forecasts with realistic uncertainty growth.
Adaptive sparseness for correntropy-based robust regression via automatic relevance determinationcs.LG
Adaptive sparseness enhances robust regression using MCC and ARD.
problem Developing a robust regression method with adaptive sparseness.
method Integrating MCC with ARD in a Bayesian framework using variational Bayesian inference.
result MCC-ARD regression outperforms existing methods in prediction and feature selection.
Credit risk modelling is an integral part of the global financial system. While there has been great attention paid to neural network models for credit default prediction, such models often lack the required interpretation mechanisms and measures of the uncertainty around their predictions. This work develops and compa…
A recurring problem when building probabilistic latent variable models is regularization and model selection, for instance, the choice of the dimensionality of the latent space. In the context of belief networks with latent variables, this problem has been adressed with Automatic Relevance Determination (ARD) employing…
We study the Automatic Relevance Determination procedure applied to deep neural networks. We show that ARD applied to Bayesian DNNs with Gaussian approximate posterior distributions leads to a variational bound similar to that of variational dropout, and in the case of a fixed dropout rate, objectives are exactly the s…
Bayesian optimisation improves with fully-Bayesian treatment of hyperparameters.
problem Overconfident model predictions in BO due to ignoring hyperparameter uncertainty.
method Investigate FBBO using three approximate inference schemes compared to maximum likelihood approach.
result FBBO using EI with an ARD kernel leads to best performance in noise-free setting.
An Automatic Relevance Determination Prior Bayesian Neural Network for Controlled Variable Selectionstat.ML
We present an Automatic Relevance Determination prior Bayesian Neural Network(BNN-ARD) weight l2-norm measure as a feature importance statistic for the model-x knockoff filter. We show on both simulated data and the Norwegian wind farm dataset that the proposed feature importance statistic yields statistically signific…
We propose a novel approach for nonlinear regression using a two-layer neural network (NN) model structure with sparsity-favoring hierarchical priors on the network weights. We present an expectation propagation (EP) approach for approximate integration over the posterior distribution of the weights, the hierarchical s…
The paper analyzes methods for sparse Bayesian regression in nonlinear system identification.
problem Learning sparse models in Bayesian regression with nonlinear applications.
method Two classes of methods: regularization and thresholding based, built on automatic relevance determination (ARD).
result Analytical demonstration of favorable performance with sparse solutions in linear problems.
Specifying utility functions is a key step towards applying the discrete choice framework for understanding the behaviour processes that govern user choices. However, identifying the utility function specifications that best model and explain the observed choices can be a very challenging and time-consuming task. This …
We propose a globally convergent alternating minimization (AM) algorithm for image reconstruction in transmission tomography, which extends automatic relevance determination (ARD) to Poisson noise models with Beer's law. The algorithm promotes solutions that are sparse in the pixel/voxel-differences domain by introduci…
NARD extends ARD for linear models, promoting sparsity and correlation structure.
problem Sparse relationships between inputs and outputs, capturing correlation structure.
method Matrix normal prior with sparsity-inducing parameter, iterative updates, sequential evaluation, and surrogate function approximation.
result Significant computational efficiency improvements with comparable performance.
Media is generally stored digitally and is therefore discrete. Many successful deep distribution models in deep learning learn a density, i.e., the distribution of a continuous random variable. Naïve optimization on discrete data leads to arbitrarily high likelihoods, and instead, it has become standard practice to add…
Jointly learns feature and sample relevancies for robust sparse recovery.
problem Sparse recovery sensitivity to data contaminants like outliers or misspecified noise.
method Jointly learns feature and sample relevancies via marginal likelihood optimization.
result Consistent sparse and robust prediction models across diverse tasks.
Financial markets are notoriously complex environments, presenting vast amounts of noisy, yet potentially informative data. We consider the problem of forecasting financial time series from a wide range of information sources using online Gaussian Processes with Automatic Relevance Determination (ARD) kernels. We measu…
New kernel interprets 3D anisotropic data with rotations and improved predictions.
problem Capturing rotated anisotropy in 3D spatial fields.
method Introduces a Lie-algebraic kernel with three principal length-scales and an explicit rotation.
result Posterior recovers rotated anisotropy and improves prediction over axis-aligned kernels.
This paper focuses on Bayesian Optimization (BO) for objectives on combinatorial search spaces, including ordinal and categorical variables. Despite the abundance of potential applications of Combinatorial BO, including chipset configuration search and neural architecture search, only a handful of methods have been pro…
RFFNet scales kernel methods to large datasets by learning kernel relevance.
problem Scaling kernel methods to large datasets while maintaining interpretability.
method Designs random Fourier features for ARD kernels and uses first-order stochastic optimization for learning kernel relevances.
result RFFNet achieves low prediction error and identifies relevant features, leading to more interpretable solutions.
The paper presents a framework for estimating treatment effects using partial network data.
problem Estimating treatment effects when interference exists and complete network data is unavailable.
method Structural causal models and various network sampling strategies.
result Validated approach using simulated experiments and real-world applications.
Develops a risk score to assist ECMO planning for critically ill patients with viral or unspecified pneumonia.
problem Lack of a risk score to guide ECMO planning for critically ill patients.
method Leverages machine learning to develop the PEER score.
result PEER score predicts mortality and decompensation in patients eligible for ECMO.
VSCOUT detects anomalies in high-dimensional data using a hybrid VAE approach.
problem Challenges in classical SPC for high-dimensional, non-Gaussian data.
method Hybrid VAE architecture with ARD prior, ensemble filtering, and changepoint detection.
result VSCOUT achieves superior sensitivity to special-cause structure and controlled false alarms.
We aim to identify the generating, ordinary differential equation (ODE) from a set of trajectories of a partially observed system. Our approach does not need prescribed basis functions to learn the ODE model, but only a rich set of Neural Arithmetic Units. For maximal explainability of the learnt model, we minimise the…
Many data-driven approaches exist to extract neural representations of functional magnetic resonance imaging (fMRI) data, but most of them lack a proper probabilistic formulation. We propose a group level scalable probabilistic sparse factor analysis (psFA) allowing spatially sparse maps, component pruning using automa…
A Neural Network for Determination of Latent Dimensionality in Nonnegative Matrix Factorizationcs.LG
A neural network method determines the latent dimensionality of NMF.
problem Determining the correct number of hidden features (latent dimensionality) in NMF.
method Combining NMFk with an MLP classifier trained on a dataset of matrices with known latent features.
result The MLP classifier in conjunction with NMFk achieves a greater than 95% success rate in determining the correct number of latent features.
A common strategy for sparse linear regression is to introduce regularization, which eliminates irrelevant features by letting the corresponding weights be zeros. However, regularization often shrinks the estimator for relevant features, which leads to incorrect feature selection. Motivated by the above-mentioned issue…
In many problem settings, parameter vectors are not merely sparse but dependent in such a way that non-zero coefficients tend to cluster together. We refer to this form of dependency as "region sparsity." Classical sparse regression methods, such as the lasso and automatic relevance determination (ARD), which model par…
The Gaussian Process Latent Variable Model (GP-LVM) is a non-linear probabilistic method of embedding a high dimensional dataset in terms low dimensional `latent' variables. In this paper we illustrate that maximum a posteriori (MAP) estimation of the latent variables and hyperparameters can be used for model selection…
GP model for time series forecasting with priors.
problem Automatic selection of optimal kernels and reliable estimation of hyperparameters.
method Fixed composition of kernels, automatic relevance determination (ARD), empirical Bayes priors.
result GP model is more accurate than state-of-the-art models.
Bayesian Tensor Ring factorization improved for scalability and handling of discrete data.
problem Scalability issues and handling of discrete data in Bayesian Tensor Ring factorization.
method Proposes a novel Bayesian Tensor Ring model with a nonparametric Multiplicative Gamma Process prior and Pólya-Gamma augmentation for discrete data. Developed efficient Gibbs sampler and online EM algorithm for scalability.
result Significantly improved scalability and handling of discrete data compared to previous methods.
BKTF uses tensor factorization for Bayesian optimization of complex functions.
problem Complex functions with nonstationary, nonseparable, and multimodal features.
method Bayesian Kernelized Tensor Factorization (BKTF) approximates complex functions using a low-rank tensor CP decomposition with GP priors.
result BKTF provides flexible and effective surrogate modeling with uncertainty quantification.
The paper proves Liouville theorems for -harmonic maps under specific curvature conditions.
problem Proving Liouville theorems for -harmonic maps in Riemannian manifolds with non-negative -Ricci curvature.
method Probabilistic proof extending previous results by Cheng, Hildebrandt-Jost-Wideman, and Stafford.
result Extends Liouville theorems to a broader class of manifolds and curvature conditions.
BONSAI optimizes parameters while respecting a default configuration, reducing unnecessary changes.
problem Standard BO pushes weakly relevant parameters to the boundary, making it hard to distinguish between important and spurious changes.
method BONSAI is a default-aware BO policy that prunes low-impact deviations from a default configuration while controlling acquisition value loss.
result BONSAI matches the GP-UCB regret rate while recovering the minimal- solution, reducing the number of non-default parameters in recommended configurations.
RFN improves GCNs for road networks, outperforming state-of-the-art by 21%-40%.
problem Leveraging the structure of road networks effectively in machine learning tasks.
method Introducing RFN, a novel GCN specifically designed for road networks.
result RFN outperforms state-of-the-art GCNs by 21%-40% on road network tasks.
This survey clarifies dynamic network terminology and reviews GNN models for dynamic networks.
problem Ambiguity in dynamic network terminology and lack of GNN models for dynamic networks.
method Established consistent terminology and notation for dynamic networks, reviewed GNN models.
result Comprehensive survey of dynamic graph neural network models.
From the perspective of network analysis, the ubiquitous networks are comprised of regular and irregular components, which makes uncovering the complexity of network structures to be a fundamental challenge. Exploring the regular information and identifying the roles of microscopic elements in network data can help us …
Deep ReLU networks can be simplified to a three-layer model.
problem Understanding the behavior of deep neural networks.
method Constructive proof and algorithm to transform deep networks into shallow ones.
result Deep ReLU networks can be represented by a simpler three-layer structure.
Chemical networks outperform spiking neural networks in classification tasks.
problem Learning tasks with spiking neural networks require hidden layers, which are computationally expensive.
method Used deterministic mass-action kinetics to prove chemical reaction networks without hidden layers can solve tasks previously solved by spiking neural networks.
result A chemical reaction network without hidden layers outperforms a spiking neural network with hidden layers in a handwritten digit classification task.
This paper explores loss landscapes of sparse neural networks, finding unique characteristics compared to dense networks.
problem Understanding the loss landscape of sparse neural networks, especially one-hidden-layer networks.
method Analyzes sparse networks with dense and sparse final layers, focusing on linear and non-linear models.
result Sparse networks can have no spurious valleys under certain conditions, but spurious valleys and minima can exist for wide sparse networks.
New approach learns latent motifs in networks for mesoscale structure analysis.
problem Understanding large-scale behavior in complex systems through mesoscale structures.
method Network dictionary learning (NDL) combining network sampling and nonnegative matrix factorization.
result Networks can be approximated using a small set of latent motifs.
A challenging problem in complex networks is the network reconstruction problem from data. This work deals with a class of networks denoted as conserved networks, in which a flow associated with every edge and the flows are conserved at all non-source and non-sink nodes. We propose a novel polynomial time algorithm to …
Social network analysis is an important problem in data mining. A fundamental step for analyzing social networks is to encode network data into low-dimensional representations, i.e., network embeddings, so that the network topology structure and other attribute information can be effectively preserved. Network represen…
Taking inspiration from biological evolution, we explore the idea of "Can deep neural networks evolve naturally over successive generations into highly efficient deep neural networks?" by introducing the notion of synthesizing new highly efficient, yet powerful deep neural networks over successive generations via an ev…
SyNGLER generates synthetic networks efficiently while preserving key structural properties.
problem Efficiently generating realistic synthetic networks with preserved structural properties.
method SyNGLER uses latent space network models to learn and reconstruct node embeddings, then generates synthetic networks.
result SyNGLER produces synthetic networks that better preserve key network characteristics than existing approaches.
Natural graph networks are a new class of graph neural networks that are more flexible and scalable.
problem Traditional graph neural networks are limited by equivariance to node permutations.
method Introduced natural graph networks, which are more flexible and scalable than conventional graph neural networks.
result Natural graph networks are as scalable as conventional message passing graph neural networks but more flexible.
The structure of complex networks has been of interest in many scientific and engineering disciplines over the decades. A number of studies in the field have been focused on finding the common properties among different kinds of networks such as heavy-tail degree distribution, small-worldness and modular structure and …
Convolutional networks outperform fully-connected ones in certain tasks.
problem Understanding the computational advantage of convolutional networks over fully-connected networks.
method Demonstrated a computational advantage through a specific problem class.
result Convolutional networks can solve certain problems that fully-connected networks cannot, even with gradient descent.