Enhances neural network regression performance by modeling weight and variance uncertainty.
problem Improving predictive performance of neural networks for regression tasks.
method Extended Blundell's framework to include variance uncertainty, using a full posterior distribution over variance parameters.
result Explicitly modeling variance uncertainty improves generalization of Bayesian neural networks.
Before training a neural net, a classic rule of thumb is to randomly initialize the weights so the variance of activations is preserved across layers. This is traditionally interpreted using the total variance due to randomness in both weights \emph{and} samples. Alternatively, one can interpret the rule of thumb as pr…
Paper improves confidence intervals and variance estimation for deep learning models.
problem Improving confidence intervals and variance estimation in deep learning models.
method Residual-based framework for conditional variance estimation; robust bootstrap procedure for confidence intervals.
result First non-asymptotic bounds for variance estimation using ReLU networks.
The bias-variance tradeoff tells us that as model complexity increases, bias falls and variances increases, leading to a U-shaped test error curve. However, recent empirical results with over-parameterized neural networks are marked by a striking absence of the classic U-shaped test error curve: test error keeps decrea…
The classical bias-variance trade-off predicts that bias decreases and variance increase with model complexity, leading to a U-shaped risk curve. Recent work calls this into question for neural networks and other over-parameterized models, for which it is often observed that larger models generalize better. We provide …
Delta Variances efficiently estimate epistemic uncertainty in neural networks.
problem Mitigating uncertainty in neural networks with limited data.
method Delta Variances, a computationally efficient algorithm for epistemic uncertainty quantification.
result Empirically competitive results with a single gradient computation.
Study on fluctuations in neural network kernels and predictions, focusing on finite width effects.
problem Characterizing fluctuations in finite width neural networks.
method Dynamical mean field theory analysis of wide but finite feature learning neural networks.
result Fluctuations in kernels and predictions are dynamically coupled, leading to reduced variance in feature learning regimes.
Paper optimizes MVE network convergence and regularization.
problem Optimizing Mean Variance Estimation networks for better performance.
method Presented two key insights: warm-up period for mean optimization and separate regularization of mean and variance.
result Warm-up period and separate regularization improve MVE network performance.
IENs reduce neural network variance without increasing complexity.
problem Reducing variance in neural networks without increasing model complexity.
method IENs use ensemble parameters during training to reduce variance, removing them during testing.
result IENs reduce network variance by a factor of 1/mL−1, leading to significant error rate decreases. This paper tackles variance issues in GNN training by proposing a method to reduce both embedding and gradient variances.
problem High variance in estimating stochastic gradients in GNN training, especially in large graphs.
method The paper proposes a decoupled variance reduction strategy that employs approximate gradient information to adaptively sample nodes with minimal variance.
result The proposed method achieves faster convergence and better generalization compared to existing sampling methods.
Optimizes MCMC chains with neural control variates.
problem Reducing variance in Markov Chain Monte Carlo (MCMC) simulations.
method Uses neural networks as control variates to minimize asymptotic variance.
result Derives optimal convergence rate under various ergodicity assumptions.
Two Fisher information matrix estimators are analyzed for neural networks, focusing on their variances and trade-offs.
problem Estimating the Fisher information matrix in neural networks due to its high computational cost.
method Examined two popular diagonal Fisher information matrix estimators and their variances in neural networks for regression and classification.
result The variances of the estimators depend on the non-linearity with respect to different parameter groups and should not be neglected.
Improved heteroscedastic regression using neural networks with provably accurate mean estimates and calibrated variance.
problem Optimizing neural network parameters for heteroscedastic regression leads to suboptimal mean and variance estimates.
method Two simple modifications to optimization to retain accuracy of mean-only models and offer best-in-class variance calibration.
result Mean estimates from the proposed method are provably as accurate as those from a homoscedastic model.
We propose and investigate new complementary methodologies for estimating predictive variance networks in regression neural networks. We derive a locally aware mini-batching scheme that result in sparse robust gradients, and show how to make unbiased weight updates to a variance network. Further, we formulate a heurist…
Study explores bias-variance in adversarial machine learning.
problem Understanding adversarial machine learning's impact on bias and variance.
method Investigates bias-variance trade-offs in deep neural networks using MSE and cross-entropy.
result Derives bias-variance trade-offs for classification and regression.
New algorithm for training GNNs with learned weights.
problem Optimal sampling for GNNs with learned weights is intractable.
method Formulated as an adversary bandit problem, optimizing exploration and exploitation.
result Asymptotically approaches optimal variance within a factor of 3.
USNRT uses tree-structured learning to improve uncertainty quantification of variance networks.
problem Improving uncertainty quantification of variance networks.
method Tree-structured local neural network model that partitions feature space into regions for training region-specific neural networks to predict mean and variance.
result USNRT shows superior performance in estimating uncertainty with variances on UCI datasets compared to recent methods.
Improves deep learning performance on noisy datasets using inverse-variance weighting.
problem Heteroscedastic regression with varying noise levels.
method Batch Inverse-Variance (BIV) loss function for neural networks.
result Significantly improves network performance on noisy datasets compared to other methods.
The main goal of this thesis is to point out that the bias-variance tradeoff is not always true (e.g. in neural networks). We advocate for this lack of universality to be acknowledged in textbooks and taught in introductory courses that cover the tradeoff. We first review the history of the bias-variance tradeoff, its …
A simple method treats heteroscedastic variance variatively, improving model calibration and sample quality.
problem Brittle optimization impacts model likelihoods for mean and variance estimation.
method Proposes a variational approach to heteroscedastic variance, improving predictive mean and variance calibration.
result The proposed method significantly improves parameter calibration and sample quality for regression and VAEs.
The study reveals a transition in neural network performance from infinite-width to variance-limited behavior as dataset size increases.
problem Understanding the transition from infinite-width to variance-limited behavior in neural networks.
method Empirical study of the transition from infinite-width to variance-limited behavior as a function of sample size and network width.
result The critical sample size \( P^* \) is approximately \( \sqrt{N} \) for polynomial regression with ReLU networks.
DN estimator mitigates network interference in experiments.
problem Network interference biases naive experiment designs.
method Differences-in-Neighbors (DN) estimator designed to mitigate interference.
result DN achieves bias second order in interference effect, with exponentially smaller variance.
The network jackknife provides conservative variance estimates for network statistics.
problem Estimating the variance of network statistics.
method Leave-node-out jackknife procedure for network data under the sparse graphon model.
result The network jackknife leads to conservative estimates of the variance for network functionals invariant to node permutation.
Stochastic neural networks with infinite width become deterministic, reducing training variance.
problem Understanding how stochasticity in neural networks affects learning and regularization.
method Theoretical analysis of stochastic neural networks with infinite width.
result As the width of an optimized stochastic neural network increases, its predictive variance on the training set decreases to zero.
Cooperative model disentangles data uncertainties.
problem Disentangling aleatoric and epistemic uncertainties in real-world data.
method Cooperatively trains a variance estimation network with a Bayesian neural network.
result Improves mean estimation and disentangles uncertainties.
Estimates generalization gap for overparameterized models using Langevin approximation.
problem Estimating the difference between training and generalization performance in overparameterized models.
method Functional variance and Langevin approximation of functional variance.
result Demonstrates efficient estimation of generalization gaps for overparameterized models.
RMDA trains structured neural networks with regularization and variance reduction.
problem Training structured neural networks with desired properties.
method RMDA algorithm for structured NNs with regularization and variance reduction.
result RMDA achieves desired structures identical to regularizer's at stationary points.
Optimizes wide low-rank neural networks for reduced parameters and cost.
problem Reducing the number of learnable parameters in wide neural networks.
method Analyzed edge-of-chaos dynamics and derived formulae for optimal weight and bias variances.
result Optimal weight and bias variances for low-rank networks follow from multiplicative scaling.
We analyze the variance of Fisher information estimators in deep learning models.
problem Understanding the variance of Fisher information in deep learning models.
method Investigated two unbiased and consistent estimators of Fisher information matrix.
result The variance of estimators is influenced by the model's parametric structure.
Bayesian neural networks (BNNs) hold great promise as a flexible and principled solution to deal with uncertainty when learning from finite data. Among approaches to realize probabilistic inference in deep neural networks, variational Bayes (VB) is theoretically grounded, generally applicable, and computationally effic…
The paper introduces a method to decompose variance in twin networks for better treatment effect estimation.
problem Accurate treatment effect estimation requires reliable uncertainty measures to locate model failures.
method Layer-wise variance decomposition using Monte Carlo Dropout in twin networks.
result The encoder component dominates under distributional shift, providing a practical diagnostic for data collection.
A method to analyze neural network performance by measuring layer saturation.
problem Understanding which layers contribute to network performance.
method Layer saturation method: restricts layer output to eigenspace of variance matrix.
result Layer saturation indicates which layers contribute to network performance.
Novel estimator reduces diffusion model variance.
problem High variance in score function estimation for diffusion models.
method Uses nearest neighbour samples to estimate the score function.
result Significant decrease in variance, leading to improved model performance.
Paper proposes a method to estimate variance reduction in DNN training using importance sampling.
problem Challenges in assessing variance reduction during DNN training using importance sampling.
method Proposes a method for estimating variance reduction using minibatches sampled under importance sampling.
result Demonstrates consistent reduction in variance, improved training efficiency, and enhanced model accuracy.
REPAIR mitigates variance collapse to enable linear interpolation between SGD solutions.
problem Linear interpolation between SGD solutions is difficult due to variance collapse in permuted activations.
method REPAIR rescales preactivations of interpolated networks to mitigate variance collapse.
result 60%-100% relative barrier reduction across various architectures and tasks.
Ordinary stochastic neural networks mostly rely on the expected values of their weights to make predictions, whereas the induced noise is mostly used to capture the uncertainty, prevent overfitting and slightly boost the performance through test-time averaging. In this paper, we introduce variance layers, a different k…
This paper develops a new portfolio optimization framework that considers network spillovers.
problem Modern financial markets' complex interconnections are not fully captured by variance alone.
method Formulates a three-objective optimization problem with a quadratic measure of network spillovers.
result Establishes a three-dimensional efficient surface and a risk-risk frontier.
DLNs dynamics change with variance, leading to saddle-to-saddle training phases.
problem Understanding the dynamics of DLNs with varying initialization variance.
method Analyzing the phase transition of DLNs' dynamics as variance changes.
result Gradient descent visits a sequence of saddles, reaching a sparse global minimum.
New algorithm reduces variance in Monte Carlo simulations using deep neural networks and policy gradients.
problem Reducing variance in Monte Carlo simulations for estimating function values.
method Optimal correlation search using deep neural networks and policy gradients.
result Optimal correlation function reduces variance by approximating and calibrating policy.
We propose a generic framework to calibrate accuracy and confidence of a prediction in deep neural networks through stochastic inferences. We interpret stochastic regularization using a Bayesian model, and analyze the relation between predictive uncertainty of networks and variance of the prediction scores obtained by …
Study on neural networks' performance under different normalizations as N grows.
problem Characterizing neural networks' performance under various normalizations.
method Developed an asymptotic expansion to analyze statistical output of shallow neural networks.
result No bias-variance trade-off exists to leading order in N, and variance decreases as normalization approaches mean field.
FGD reduces noisy gradient variance in SGD for neural networks.
problem Noisy and unreliable gradient estimation in SGD for deep learning.
method Solves an adaptive filtering problem to consistently estimate the local gradient.
result Significantly reduces gradient variance and accelerates convergence.
A new aggregation strategy improves GNN performance and learning dynamics.
problem Improving expressivity and learning dynamics of GNNs.
method Proposes a variance-preserving aggregation function (VPA) for GNNs.
result VPA leads to increased predictive performance and improved learning dynamics.
MARS optimizes large model training by reducing variance, outperforming AdamW.
problem Training large models efficiently and scalably.
method Unified optimization framework MARS combining preconditioned gradient updates and variance reduction.
result MARS outperforms AdamW in training GPT-2 models.
Proposes deep graph persistence to address neural persistence issues in deep learning.
problem Variance of weights and lack of spatial structure in deep neural networks impact neural persistence.
method Extends neural persistence to the whole network, considering interactions between layers.
result Deep graph persistence alleviates variance-related issues and captures persistent paths through the network.
VAEs analyzed using harmonic analysis, showing how variance controls frequency content and robustness.
problem Understanding and optimizing VAEs for robustness and frequency control.
method Viewing VAE latent space as Gaussian space, deriving results on variance and frequency content, and demonstrating soft Lipschitz constraints.
result Increasing encoder variance reduces high frequency content and improves adversarial robustness.
New methods solve graph sparsity optimization problems faster.
problem Complex graph sparsity optimization problems in disease outbreak monitoring and social network analysis.
method Stochastic variance-reduced gradient-based methods GraphSVRG-IHT and GraphSCSG-IHT.
result Our methods achieve linear convergence speed.
Improves early stopping in deep networks by adjusting stepsizes.
problem Epoch-wise double descent in deep networks.
method Analytical and empirical study of bias-variance tradeoffs in different network layers.
result Eliminating epoch-wise double descent through adjusting stepsizes of different layers improves early stopping performance.