Search Results for author: Willem Waegeman

Found 23 papers, 7 papers with code

Is Epistemic Uncertainty Faithfully Represented by Evidential Deep Learning Methods?

no code implementations14 Feb 2024 Mira Jürgens, Nis Meinert, Viktor Bengs, Eyke Hüllermeier, Willem Waegeman

Trustworthy ML systems should not only return accurate predictions, but also a reliable representation of their uncertainty.

Heteroskedastic conformal regression

1 code implementation15 Sep 2023 Nicolas Dewolf, Bernard De Baets, Willem Waegeman

Conformal prediction, and split conformal prediction as a specific implementation, offer a distribution-free approach to estimating prediction intervals with statistical guarantees.

Conformal Prediction Prediction Intervals +1

The out-of-sample $R^2$: estimation and inference

1 code implementation10 Feb 2023 Stijn Hawinkel, Willem Waegeman, Steven Maere

Out-of-sample prediction is the acid test of predictive models, yet an independent test dataset is often not available for assessment of the prediction error.

On Second-Order Scoring Rules for Epistemic Uncertainty Quantification

no code implementations30 Jan 2023 Viktor Bengs, Eyke Hüllermeier, Willem Waegeman

In this paper, we generalise these findings and prove a more fundamental result: There seems to be no loss function that provides an incentive for a second-order learner to faithfully represent its epistemic uncertainty in the same manner as proper scoring rules do for standard (first-order) learners.

Uncertainty Quantification

Hyperparameter optimization in deep multi-target prediction

1 code implementation8 Nov 2022 Dimitrios Iliadis, Marcel Wever, Bernard De Baets, Willem Waegeman

As a result of the ever increasing complexity of configuring and fine-tuning machine learning models, the field of automated machine learning (AutoML) has emerged over the past decade.

Benchmarking Hyperparameter Optimization +5

On the Calibration of Probabilistic Classifier Sets

no code implementations20 May 2022 Thomas Mortier, Viktor Bengs, Eyke Hüllermeier, Stijn Luca, Willem Waegeman

In this paper, we extend the notion of calibration, which is commonly used to evaluate the validity of the aleatoric uncertainty representation of a single probabilistic classifier, to assess the validity of an epistemic uncertainty representation obtained by sets of probabilistic classifiers.

Ensemble Learning Multi-class Classification

Pitfalls of Epistemic Uncertainty Quantification through Loss Minimisation

no code implementations11 Mar 2022 Viktor Bengs, Eyke Hüllermeier, Willem Waegeman

Uncertainty quantification has received increasing attention in machine learning in the recent past.

Uncertainty Quantification

Valid prediction intervals for regression problems

1 code implementation1 Jul 2021 Nicolas Dewolf, Bernard De Baets, Willem Waegeman

Over the last few decades, various methods have been proposed for estimating prediction intervals in regression settings, including Bayesian methods, ensemble methods, direct interval estimation methods and conformal prediction methods.

Conformal Prediction Prediction Intervals +2

Multi-target prediction for dummies using two-branch neural networks

no code implementations19 Apr 2021 Dimitrios Iliadis, Bernard De Baets, Willem Waegeman

In this work we present a generic deep learning methodology that can be used for a wide range of multi-target prediction problems.

BIG-bench Machine Learning Matrix Completion +5

Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methods

1 code implementation21 Oct 2019 Eyke Hüllermeier, Willem Waegeman

The notion of uncertainty is of major importance in machine learning and constitutes a key element of machine learning methodology.

BIG-bench Machine Learning

Efficient Set-Valued Prediction in Multi-Class Classification

4 code implementations19 Jun 2019 Thomas Mortier, Marek Wydmuch, Krzysztof Dembczyński, Eyke Hüllermeier, Willem Waegeman

In cases of uncertainty, a multi-class classifier preferably returns a set of candidate classes instead of predicting a single class label with little guarantee.

Classification General Classification +1

Multi-Target Prediction: A Unifying View on Problems and Methods

no code implementations7 Sep 2018 Willem Waegeman, Krzysztof Dembczynski, Eyke Huellermeier

Multi-target prediction (MTP) is concerned with the simultaneous prediction of multiple target variables of diverse type.

Matrix Completion Multi-Label Classification +2

Exact and efficient top-K inference for multi-target prediction by querying separable linear relational models

no code implementations14 Jun 2016 Michiel Stock, Krzysztof Dembczynski, Bernard De Baets, Willem Waegeman

Many complex multi-target prediction problems that concern large target spaces are characterised by a need for efficient prediction strategies that avoid the computation of predictions for all targets explicitly.

BIG-bench Machine Learning Collaborative Filtering +3

Efficient Pairwise Learning Using Kernel Ridge Regression: an Exact Two-Step Method

no code implementations14 Jun 2016 Michiel Stock, Tapio Pahikkala, Antti Airola, Bernard De Baets, Willem Waegeman

In this work we analyze kernel-based methods for pairwise learning, with a particular focus on a recently-suggested two-step method.

Collaborative Filtering Matrix Completion +3

Spectral Analysis of Symmetric and Anti-Symmetric Pairwise Kernels

no code implementations19 Jun 2015 Tapio Pahikkala, Markus Viljanen, Antti Airola, Willem Waegeman

We consider the problem of learning regression functions from pairwise data when there exists prior knowledge that the relation to be learned is symmetric or anti-symmetric.

regression

Identification of functionally related enzymes by learning-to-rank methods

no code implementations17 May 2014 Michiel Stock, Thomas Fober, Eyke Hüllermeier, Serghei Glinca, Gerhard Klebe, Tapio Pahikkala, Antti Airola, Bernard De Baets, Willem Waegeman

For a given query, the search operation results in a ranking of the enzymes in the database, from very similar to dissimilar enzymes, while information about the biological function of annotated database enzymes is ignored.

Learning-To-Rank

On the Bayes-optimality of F-measure maximizers

no code implementations17 Oct 2013 Willem Waegeman, Krzysztof Dembczynski, Arkadiusz Jachnik, Weiwei Cheng, Eyke Hullermeier

The F-measure, which has originally been introduced in information retrieval, is nowadays routinely used as a performance metric for problems such as binary classification, multi-label classification, and structured output prediction.

Binary Classification General Classification +3

Efficient Regularized Least-Squares Algorithms for Conditional Ranking on Relational Data

no code implementations21 Sep 2012 Tapio Pahikkala, Antti Airola, Michiel Stock, Bernard De Baets, Willem Waegeman

In domains like bioinformatics, information retrieval and social network analysis, one can find learning tasks where the goal consists of inferring a ranking of objects, conditioned on a particular target object.

Computational Efficiency Information Retrieval +2

An Exact Algorithm for F-Measure Maximization

no code implementations NeurIPS 2011 Krzysztof J. Dembczynski, Willem Waegeman, Weiwei Cheng, Eyke Hüllermeier

The F-measure, originally introduced in information retrieval, is nowadays routinely used as a performance metric for problems such as binary classification, multi-label classification, and structured output prediction.

Binary Classification Classification +4

Cannot find the paper you are looking for? You can Submit a new open access paper.