Search Results for author:
Found 44 papers, 23 papers with code
On Calibration in Multi-Distribution Learning
Modern challenges of robustness, fairness, and decision-making in machine learning have led to the formulation of multi-distribution learning (MDL) frameworks in which a predictor is optimized across multiple distributions.
ELBOing Stein: Variational Bayes with Stein Mixture Inference
Stein variational gradient descent (SVGD) [Liu and Wang, 2016] performs approximate Bayesian inference by representing the posterior with a set of particles.
DefVerify: Do Hate Speech Models Reflect Their Dataset's Definition?
To address this issue for hate speech detection, we propose DefVerify: a 3-step procedure that (i) encodes a user-specified definition of hate speech, (ii) quantifies to what extent the model reflects the intended definition, and (iii) tries to identify the point of failure in the workflow.
Lightning UQ Box: A Comprehensive Framework for Uncertainty Quantification in Deep Learning
Uncertainty quantification (UQ) is an essential tool for applying deep neural networks (DNNs) to real world tasks, as it attaches a degree of confidence to DNN outputs.
Crowd-Calibrator: Can Annotator Disagreement Inform Calibration in Subjective Tasks?
Subjective tasks in NLP have been mostly relegated to objective standards, where the gold label is decided by taking the majority vote.
Temporal Test-Time Adaptation with State-Space Models
Distribution shifts between training and test data are inevitable over the lifecycle of a deployed model, leading to performance decay.
Fast yet Safe: Early-Exiting with Risk Control
Scaling machine learning models significantly improves their performance.
Uncertainty Aware Tropical Cyclone Wind Speed Estimation from Satellite Data
We provide a detailed evaluation of predictive uncertainty estimates from state-of-the-art uncertainty quantification (UQ) methods for DNNs.
Adaptive Bounding Box Uncertainties via Two-Step Conformal Prediction
Thus, we develop a novel two-step conformal approach that propagates uncertainty in predicted class labels into the uncertainty intervals of bounding boxes.
Learning to Defer to a Population: A Meta-Learning Approach
The learning to defer (L2D) framework allows autonomous systems to be safe and robust by allocating difficult decisions to a human expert.
A Generative Model of Symmetry Transformations
Correctly capturing the symmetry transformations of data can lead to efficient models with strong generalization capabilities, though methods incorporating symmetries often require prior knowledge.
On the Challenges and Opportunities in Generative AI
The field of deep generative modeling has grown rapidly and consistently over the years.
Beyond Top-Class Agreement: Using Divergences to Forecast Performance under Distribution Shift
Knowing if a model will generalize to data 'in the wild' is crucial for safe deployment.
One Strike, You're Out: Detecting Markush Structures in Low Signal-to-Noise Ratio Images
The focus of this research was to propose and test a novel method for classifying Markush structures.
Max-Rank: Efficient Multiple Testing for Conformal Prediction
Multiple hypothesis testing (MHT) commonly arises in various scientific fields, from genomics to psychology, where testing many hypotheses simultaneously increases the risk of Type-I errors.
Early-Exit Neural Networks with Nested Prediction Sets
These sequences are inherently nested and thus well-suited for an EENN's sequential predictions.
Active Learning for Multilingual Fingerspelling Corpora
We apply active learning to help with data scarcity problems in sign languages.
Exploiting Inferential Structure in Neural Processes
Neural Processes (NPs) are appealing due to their ability to perform fast adaptation based on a context set.
Do Bayesian Neural Networks Need To Be Fully Stochastic?
We investigate the benefit of treating all the parameters in a Bayesian neural network stochastically and find compelling theoretical and empirical evidence that this standard construction may be unnecessary.
Learning to Defer to Multiple Experts: Consistent Surrogate Losses, Confidence Calibration, and Conformal Ensembles
We study the statistical properties of learning to defer (L2D) to multiple experts.
Sampling-based inference for large linear models, with application to linearised Laplace
Large-scale linear models are ubiquitous throughout machine learning, with contemporary application as surrogate models for neural network uncertainty quantification; that is, the linearised Laplace method.
Hate Speech Criteria: A Modular Approach to Task-Specific Hate Speech Definitions
We argue that the goal and exact task developers have in mind should determine how the scope of \textit{hate speech} is defined.
Adapting the Linearised Laplace Model Evidence for Modern Deep Learning
The linearised Laplace method for estimating model uncertainty has received renewed attention in the Bayesian deep learning community.
Adversarial Defense via Image Denoising with Chaotic Encryption
In the literature on adversarial examples, white box and black box attacks have received the most attention.
Calibrated Learning to Defer with One-vs-All Classifiers
We find that Mozannar & Sontag's (2020) multiclass framework is not calibrated with respect to expert correctness.
Linearised Laplace Inference in Networks with Normalisation Layers and the Neural g-Prior
We show that for neural networks (NN) with normalisation layers, i. e. batch norm, layer norm, or group norm, the Laplace model evidence does not approximate the volume of a posterior mode and is thus unsuitable for model selection.
How Emotionally Stable is ALBERT? Testing Robustness with Stochastic Weight Averaging on a Sentiment Analysis Task
Despite their success, modern language models are fragile.
On the Inconsistency of Bayesian Inference for Misspecified Neural Networks
Grunwald and Van Ommen (2017) show that Bayesian inference for linear regression can be inconsistent under model misspecification.
Bayesian Deep Learning via Subnetwork Inference
In particular, we implement subnetwork linearized Laplace as a simple, scalable Bayesian deep learning method: We first obtain a MAP estimate of all weights and then infer a full-covariance Gaussian posterior over a subnetwork using the linearized Laplace approximation.
Expressive yet Tractable Bayesian Deep Learning via Subnetwork Inference
In particular, we develop a practical and scalable Bayesian deep learning method that first trains a point estimate, and then infers a full covariance Gaussian posterior approximation over a subnetwork.
Predictive Complexity Priors
For this reason, we propose predictive complexity priors: a functional prior that is defined by comparing the model's predictions to those of a reference model.
Normalizing Flows for Probabilistic Modeling and Inference
In this review, we attempt to provide such a perspective by describing flows through the lens of probabilistic modeling and inference.
Bayesian Batch Active Learning as Sparse Subset Approximation
Leveraging the wealth of unlabeled data produced in recent years provides great potential for improving supervised models.
Detecting Out-of-Distribution Inputs to Deep Generative Models Using Typicality
To determine whether or not inputs reside in the typical set, we propose a statistically principled, easy-to-implement test using the empirical distribution of model likelihoods.
Hybrid Models with Deep and Invertible Features
We propose a neural hybrid model consisting of a linear model defined on a set of features computed by a deep, invertible transformation (i. e. a normalizing flow).
Do Deep Generative Models Know What They Don't Know?
A neural network deployed in the wild may be asked to make predictions for inputs that were drawn from a different distribution than that of the training data.
Dropout as a Structured Shrinkage Prior
We propose a novel framework for understanding multiplicative noise in neural networks, considering continuous distributions as well as Bernoulli noise (i. e. dropout).
THE EFFECTIVENESS OF A TWO-LAYER NEURAL NETWORK FOR RECOMMENDATIONS
We present a personalized recommender system using neural network for recommending products, such as eBooks, audio-books, Mobile Apps, Video and Music.
Mondrian Processes for Flow Cytometry Analysis
Analysis of flow cytometry data is an essential tool for clinical diagnosis of hematological and immunological conditions.
Learning Approximately Objective Priors
Informative Bayesian priors are often difficult to elicit, and when this is the case, modelers usually turn to noninformative or objective priors.
Stick-Breaking Variational Autoencoders
We extend Stochastic Gradient Variational Bayes to perform posterior inference for the weights of Stick-Breaking processes.
A Dual Embedding Space Model for Document Ranking
A fundamental goal of search engines is to identify, given a query, documents that have relevant text.
Learning the Dimensionality of Word Embeddings
We describe a method for learning word embeddings with data-dependent dimensionality.
A Scale Mixture Perspective of Multiplicative Noise in Neural Networks
Corrupting the input and hidden layers of deep neural networks (DNNs) with multiplicative noise, often drawn from the Bernoulli distribution (or 'dropout'), provides regularization that has significantly contributed to deep learning's success.
