Dirk Thierens

Arthur Guijt, Dirk Thierens, Ellen Kerkhof et al.

Deep learning has been shown to be very capable at performing many real-world tasks. However, this performance is often dependent on the presence of large and varied datasets. In some settings, like in the medical domain, data is often fragmented across parties, and cannot be readily shared. While federated learning addresses this situation, it is a solution that requires synchronicity of parties training a single model together, exchanging information about model weights. We investigate how asynchronous collaboration, where only already trained models are shared (e.g. as part of a publication), affects performance, and propose to use stitching as a method for combining models. Through taking a multi-objective perspective, where performance on each parties' data is viewed independently, we find that training solely on a single parties' data results in similar performance when merging with another parties' data, when considering performance on that single parties' data, while performance on other parties' data is notably worse. Moreover, while an ensemble of such individually trained networks generalizes better, performance on each parties' own dataset suffers. We find that combining intermediate representations in individually trained models with a well placed pair of stitching layers allows this performance to recover to a competitive degree while maintaining improved generalization, showing that asynchronous collaboration can yield competitive results.

Sjoerd de Vries, Dirk Thierens

In supervised machine learning, models are typically trained using data with hard labels, i.e., definite assignments of class membership. This traditional approach, however, does not take the inherent uncertainty in these labels into account. We investigate whether incorporating label uncertainty, represented as discrete probability distributions over the class labels -- known as soft labels -- improves the predictive performance of classification models. We first demonstrate the potential value of soft label learning (SLL) for estimating model parameters in a simulation experiment, particularly for limited sample sizes and imbalanced data. Subsequently, we compare the performance of various wrapper methods for learning from both hard and soft labels using identical base classifiers. On real-world-inspired synthetic data with clean labels, the SLL methods consistently outperform hard label methods. Since real-world data is often noisy and precise soft labels are challenging to obtain, we study the effect that noisy probability estimates have on model performance. Alongside conventional noise models, our study examines four types of miscalibration that are known to affect human annotators. The results show that SLL methods outperform the hard label methods in the majority of settings. Finally, we evaluate the methods on a real-world dataset with confidence scores, where the SLL methods are shown to match the traditional methods for predicting the (noisy) hard labels while providing more accurate confidence estimates.

Sjoerd de Vries, Dirk Thierens

In many real-world classification tasks, label noise is an unavoidable issue that adversely affects the generalization error of machine learning models. Additionally, evaluating how methods handle such noise is complicated, as the effect label noise has on their performance cannot be accurately quantified without clean labels. Existing research on label noise typically relies on either noisy or oversimplified simulated data as a baseline, into which additional noise with known properties is injected. In this paper, we introduce SYNLABEL, a framework designed to address these limitations by creating noiseless datasets informed by real-world data. SYNLABEL supports defining a pre-specified or learned function as the ground truth function, which can then be used for generating new clean labels. Furthermore, by repeatedly resampling values for selected features within the domain of the function, evaluating the function and aggregating the resulting labels, each data point can be assigned a soft label or label distribution. These distributions capture the inherent uncertainty present in many real-world datasets and enable the direct injection and quantification of label noise. The generated datasets serve as a clean baseline of adjustable complexity, into which various types of noise can be introduced. Additionally, they facilitate research into soft label learning and related applications. We demonstrate the application of SYNLABEL, showcasing its ability to precisely quantify label noise and its improvement over existing methodologies.

Arkadiy Dushatskiy, Marco Virgolin, Anton Bouter et al.

When it comes to solving optimization problems with evolutionary algorithms (EAs) in a reliable and scalable manner, detecting and exploiting linkage information, i.e., dependencies between variables, can be key. In this article, we present the latest version of, and propose substantial enhancements to, the Gene-pool Optimal Mixing Evoutionary Algorithm (GOMEA): an EA explicitly designed to estimate and exploit linkage information. We begin by performing a large-scale search over several GOMEA design choices, to understand what matters most and obtain a generally best-performing version of the algorithm. Next, we introduce a novel version of GOMEA, called CGOMEA, where linkage-based variation is further improved by filtering solution mating based on conditional dependencies. We compare our latest version of GOMEA, the newly introduced CGOMEA, and another contending linkage-aware EA DSMGA-II in an extensive experimental evaluation, involving a benchmark set of 9 black-box problems that can only be solved efficiently if their inherent dependency structure is unveiled and exploited. Finally, in an attempt to make EAs more usable and resilient to parameter choices, we investigate the performance of different automatic population management schemes for GOMEA and CGOMEA, de facto making the EAs parameterless. Our results show that GOMEA and CGOMEA significantly outperform the original GOMEA and DSMGA-II on most problems, setting a new state of the art for the field.