arXiv research

Hybrid Bayesian neural networks use function uncertainty for probabilistic inference.

We describe Bayesian Layers, a module designed for fast experimentation with neural network uncertainty. It extends neural network libraries with drop-in replacements for common layers. This enables composition via a unified abstraction over deterministic and stochastic functions and allows for scalability via the unde…

Paper introduces probabilistic approach to CO layers in ML.

A new neural network model MDRBM improves noise-robustness in classification.

New CNN approach reduces overconfidence in object classification predictions.

Probabilistic graphical models are traditionally known for their successes in generative modeling. In this work, we advocate layered graphical models (LGMs) for probabilistic discriminative learning. To this end, we design LGMs in close analogy to neural networks (NNs), that is, they have deep hierarchical structures a…

ProFITi model forecasts irregular time series with missing values using conditional flows.

A generative Bayesian model is developed for deep (multi-layer) convolutional dictionary learning. A novel probabilistic pooling operation is integrated into the deep model, yielding efficient bottom-up and top-down probabilistic learning. After learning the deep convolutional dictionary, testing is implemented via dec…

A generative model is developed for deep (multi-layered) convolutional dictionary learning. A novel probabilistic pooling operation is integrated into the deep model, yielding efficient bottom-up (pretraining) and top-down (refinement) probabilistic learning. Experimental results demonstrate powerful capabilities of th…

EMFs combine deep learning and probabilistic models for better density estimation.

Study on rich regime training in deep learning, finding active parameters in bottom layers.

Deep learning uses layers of transformations to predict structured data with uncertainty.

Study on deep neural networks using concentration inequalities and optimal stopping.

Probabilistic programming aids in automatically dating ice cores, reducing manual error and uncertainty.

In this work, we introduce a novel probabilistic representation of deep learning, which provides an explicit explanation for the Deep Neural Networks (DNNs) in three aspects: (i) neurons define the energy of a Gibbs distribution; (ii) the hidden layers of DNNs formulate Gibbs distributions; and (iii) the whole architec…

A new probabilistic mixup framework improves deep learning generalization.

A new beta-VAE based regression model accelerates oilfield optimization studies.

Proposes a deep probabilistic multi-view model for multi-view learning.

We study two-layer belief networks of binary random variables in which the conditional probabilities Pr[childlparents] depend monotonically on weighted sums of the parents. In large networks where exact probabilistic inference is intractable, we show how to compute upper and lower bounds on many probabilities of intere…

The paper provides a uniform convergence bound for smooth calibration error and its relationship with functional gradient.

PHP connects to ReLU neural networks for scalable Bayesian inference.

In this paper, we propose a probabilistic parsing model, which defines a proper conditional probability distribution over non-projective dependency trees for a given sentence, using neural representations as inputs. The neural network architecture is based on bi-directional LSTM-CNNs which benefits from both word- and …

We give a polynomial-time algorithm for learning neural networks with one layer of sigmoids feeding into any Lipschitz, monotone activation function (e.g., sigmoid or ReLU). We make no assumptions on the structure of the network, and the algorithm succeeds with respect to {\em any} distribution on the unit ball in $n$ …

SVGP KAN integrates sparse variational GP with KANs for scalable probabilistic inference.

Probabilistic deep learning uses neural networks and models to handle uncertainty.

Proposes PRMs for interpreting financial risk concept drift.

Study introduces a probabilistic framework for air-sea fluxes using neural networks.

We establish $L^{\infty}$ and $L^2$ error bounds for functions of many variables that are approximated by linear combinations of ReLU (rectified linear unit) and squared ReLU ridge functions with $\ell^1$ and $\ell^0$ controls on their inner and outer parameters. With the squared ReLU ridge function, we show th…

We propose the Limited Multi-Label (LML) projection layer as a new primitive operation for end-to-end learning systems. The LML layer provides a probabilistic way of modeling multi-label predictions limited to having exactly k labels. We derive efficient forward and backward passes for this layer and show how the layer…

Scaling ResNets requires careful consideration of the layer depth and output scaling factors.

Paper analyzes GCNN sensitivity to probabilistic graph perturbations.

Low bit-width weights and activations are an effective way of combating the increasing need for both memory and compute power of Deep Neural Networks. In this work, we present a probabilistic training method for Neural Network with both binary weights and activations, called BLRNet. By embracing stochasticity during tr…

We introduce and demonstrate the variational autoencoder (VAE) for probabilistic non-negative matrix factorisation (PAE-NMF). We design a network which can perform non-negative matrix factorisation (NMF) and add in aspects of a VAE to make the coefficients of the latent space probabilistic. By restricting the weights i…

Deep models predict intraday electricity prices accurately.

Hierarchical probabilistic models are able to use a large number of parameters to create a model with a high representation power. However, it is well known that increasing the number of parameters also increases the complexity of the model which leads to a bias-variance trade-off. Although it is a classical problem, t…

K-DAREK improves KKANs for efficient function approximation with robust error bounds.

Deep Gaussian processes (DGPs) are multi-layer hierarchical generalisations of Gaussian processes (GPs) and are formally equivalent to neural networks with multiple, infinitely wide hidden layers. DGPs are probabilistic and non-parametric and as such are arguably more flexible, have a greater capacity to generalise, an…

Combines deep and statistical learning for structured data.

A new method for Bayesian neural networks using probabilistic backpropagation.

Consider a feedforward neural network $ψ: \mathbb{R}^d\rightarrow \mathbb{R}^d$ such that $ψ\approx \nabla f$, where $f:\mathbb{R}^d \rightarrow \mathbb{R}$ is a smooth function, therefore $ψ$ must satisfy $\partial_j ψ_i = \partial_i ψ_j$ pointwise. We prove a theorem that a $ψ$ network with more than one hidden layer…

New insights into optimizing latent representations in hierarchical VAEs.

DeepKriging uses neural networks for spatio-temporal interpolation and forecasting.

This paper introduces a probabilistic framework for k-shot image classification. The goal is to generalise from an initial large-scale classification task to a separate task comprising new classes and small numbers of examples. The new approach not only leverages the feature-based representation learned by a neural net…

Proposes PSCs for UQ in deep nets without retraining.

The Softmax function is used in the final layer of nearly all existing sequence-to-sequence models for language generation. However, it is usually the slowest layer to compute which limits the vocabulary size to a subset of most frequent types; and it has a large memory footprint. We propose a general technique for rep…

INNs can approximate diverse functions despite layer restrictions.

Normalizing flows are shown to be equivalent to Bayesian networks, revealing new insights.

Quantitatively assessing relationships between latent variables and observed variables is important for understanding and developing generative models and representation learning. In this paper, we propose latent-observed dissimilarity (LOD) to evaluate the dissimilarity between the probabilistic characteristics of lat…