arXiv research

Enhances neural network regression performance by modeling weight and variance uncertainty.

A method to reduce memory usage in deep learning models by adding inducing weights.

Learning probability distributions on the weights of neural networks (NNs) has recently proven beneficial in many applications. Bayesian methods, such as Stein variational gradient descent (SVGD), offer an elegant framework to reason about NN model uncertainty. However, by assuming independent Gaussian priors for the i…

SCS identifies a range of plausible equally weighted portfolios, quantifying selection uncertainty.

We introduce a new, efficient, principled and backpropagation-compatible algorithm for learning a probability distribution on the weights of a neural network, called Bayes by Backprop. It regularises the weights by minimising a compression cost, known as the variational free energy or the expected lower bound on the ma…

A new method estimates uncertainty without explicit prediction models.

Proposes a new method for uncertainty estimation in neural networks.

Enhances reinforcement learning uncertainty estimation with a generalized Gaussian error model.

New method reduces model bias and variance by adjusting training sample weights based on label uncertainty.

We introduce a novel framework for the estimation of the posterior distribution over the weights of a neural network, based on a new probabilistic interpretation of adaptive optimisation algorithms such as AdaGrad and Adam. We demonstrate the effectiveness of our Bayesian Adam method, Badam, by experimentally showing t…

This paper introduces hierarchical Gaussian process priors for neural networks to capture weight correlations and inductive biases.

Proposes a weighted conformal approach for cluster label uncertainty.

New algorithm reduces regret in RL with adversarial corruption.

The paper tackles uncertainty quantification in multi-source settings.

Study on Bayesian transformers finds issues with weight-space inference and prior specification.

Bayesian Neural Networks improve geophysical model ensembles with reduced uncertainty.

Proposes a method to quantify uncertainty in predictions under covariate shift.

PROWL uses robust reward estimates to improve ITR selection.

Hybrid Bayesian-conformal framework improves uncertainty quantification in healthcare predictions.

Extend CPS to non-exchangeable settings with observation-specific permutation weights

Uncertainty computation in deep learning is essential to design robust and reliable systems. Variational inference (VI) is a promising approach for such computation, but requires more effort to implement and execute compared to maximum-likelihood methods. In this paper, we propose new natural-gradient algorithms to red…

Survey and framework for consistent uncertainty quantification in deep learning.

Deep Reinforcement Learning improves with Weighted Q-Learning to reduce bias and uncertainty.

Hybrid Bayesian neural networks use function uncertainty for probabilistic inference.

Paper presents a method to summarize HMC samples for neural networks, providing meaningful uncertainty estimates.

This note explores probabilistic sampling weighted by uncertainty in active learning. This method has been previously used and authors have tangentially remarked on its efficacy. The scheme has several benefits: (1) it is computationally cheap, (2) it can be implemented in a single-pass streaming fashion which is a ben…

R2D2-Net improves Bayesian neural networks by preventing over-shrinkage of important weights.

We consider a setting where an agent's uncertainty is represented by a set of probability measures, rather than a single measure. Measure-bymeasure updating of such a set of measures upon acquiring new information is well-known to suffer from problems; agents are not always able to learn appropriately. To deal with the…

This study improves uncertainty quantification in seismic inversion.

The study shows how probability weighting can lead to betting in a risk-averse economy.

Teaches uncertainty in ML through practical examples.

During training, the weights of a Deep Neural Network (DNN) are optimized from a random initialization towards a nearly optimum value minimizing a loss function. Only this final state of the weights is typically kept for testing, while the wealth of information on the geometry of the weight space, accumulated over the …

Paper develops robust $k$-NN algorithm for few samples.

Paper introduces new risk measures for default risk and model uncertainty.

TFB simplifies Bayesian LLM uncertainty estimation without extra training.

New active learning methods for Gaussian process improve predictive modeling of composite fuselage.

We improve prediction set coverage by assigning weights to individual sets.

This paper benchmarks uncertainty disentanglement across various tasks.

AlphaNet improves supernets training with alpha-divergence.

Gaussian Processes improve data interpolation from diverse experiments.

As machine learning systems get widely adopted for high-stake decisions, quantifying uncertainty over predictions becomes crucial. While modern neural networks are making remarkable gains in terms of predictive accuracy, characterizing uncertainty over the parameters of these models is challenging because of the high d…

A new method forecasts financial tail risks by combining and weighting quantiles.

Long Short-Term Memory networks trained with gradient descent and back-propagation have received great success in various applications. However, point estimation of the weights of the networks is prone to over-fitting problems and lacks important uncertainty information associated with the estimation. However, exact Ba…

Unified framework for reliable uncertainty quantification in RL.

Optimizes sampling for faster convergence in Bayesian experimental design and uncertainty quantification.

Continual learning aims to learn new tasks without forgetting previously learned ones. This is especially challenging when one cannot access data from previous tasks and when the model has a fixed capacity. Current regularization-based continual learning algorithms need an external representation and extra computation …

The aggregation of k-ary preferences is a historical and important problem, since it has many real-world applications, such as peer grading, presidential elections and restaurant ranking. Meanwhile, variants of Plackett-Luce model has been applied to aggregate k-ary preferences. However, there are two urgent issues sti…

A novel method to propagate uncertainty through the soft-thresholding nonlinearity is proposed in this paper. At every layer the current distribution of the target vector is represented as a spike and slab distribution, which represents the probabilities of each variable being zero, or Gaussian-distributed. Using the p…