arXiv subjects

Thomas Trappenberg

Publications and source records attributed to Thomas Trappenberg.

At least 19 recordsLinked to original sources

arXiv2026

Uncertainty quantification for trustworthy deep learning: Methods and measures

The deployment of deep neural networks in safety-critical domains demands reliable estimates of predictive confidence, yet conventional architectures lack principled uncertainty quantification. This survey provides a structured, critical review of methods for Uncertainty Quantification (UQ) in deep learning, scoped to ensemble-based and approximate Bayesian approaches and the measures used to summarize their outputs. Relative to existing UQ surveys, our contribution is depth on efficient ensemble approximations and single-pass methods, and a unified treatment that separates the method producing a predictive distribution from the measure that summarizes its uncertainty. We organize methods into five families: Bayesian neural networks, Monte Carlo Dropout, deep ensembles, efficient ensemble approximations, and last-layer or single-pass approaches. We situate adjacent work on evidential and prior networks, conformal prediction, and post-hoc calibration, together with the decision-time tasks of out-of-distribution detection and selective prediction. For each, we examine theoretical motivation, implementation, empirical performance, and limitations. We then review ensemble diversity theory and uncertainty measures and their decompositions, contrasting the entropy decomposition with pairwise divergence measures, and consolidate evaluation methodology so that our qualitative comparisons share a common basis. We close with a brief treatment of uncertainty in large language models and open research directions, including efficient epistemic measures for classification, last-layer diversity, diversity and calibration under shift, and hybrid architectures.

stat.ML↗

arXiv2026

Covariance Last-Layer Ensembles: Function-Space Diversity for Efficient Uncertainty Quantification

A Last-Layer Ensemble (LLE), $K$ linear units on one shared frozen feature map, is an efficient single-pass approach to the disagreement-based epistemic uncertainty for out-of-distribution (OOD) detection. Its weakness is that members share the backbone gradient and can converge toward the same function, collapsing the inter-member diversity the signal depends on. Whether last-layer diversity can be restored, and what mitigates the collapse, is an open question. The weight-orthonormality defining Orthonormal Certificates (OC), the weight-orthonormal special case of the LLE, is only an indirect correction; it decorrelates the weights of the members, not their predictions. Here, we instead target the collapse directly in function space, with a Covariance Last-Layer Ensemble (cov-LLE) that places a direct covariance penalty on member activations. Cov-LLE restores the function-space diversity that weight-orthonormality cannot, and at matched $K$ recovers much of the diversity and calibration of a deep ensemble at $1\times$ backbone cost (in-distribution prediction variance $0.05\!\to\!9.3$ vs. $22.1$ ($\times10^{-3}$), and ECE $0.135\!\to\!0.090$ vs. $0.035$, for a $K\times$-cost deep ensemble), at no cost to accuracy. Viewing OC as a last-layer ensemble also organizes detectors into a two-axis taxonomy (by how their units are trained and how their outputs are scored) and exposes the OC score as a magnitude, motivating a scale-invariant, label-free direction score that repairs its near-OOD failure, adding $+0.16$ to $+0.18$ ROC AUC on every backbone.

cs.LG↗

arXiv2026

Improving Detection of Rare Nodes in Hierarchical Multi-Label Learning

In hierarchical multi-label classification, a persistent challenge is enabling model predictions to reach deeper levels of the hierarchy for more detailed or fine-grained classifications. This difficulty partly arises from the natural rarity of certain classes (or hierarchical nodes) and the hierarchical constraint that ensures child nodes are almost always less frequent than their parents. To address this, we propose a weighted loss objective for neural networks that combines node-wise imbalance weighting with focal weighting components, the latter leveraging modern quantification of ensemble uncertainties. By emphasizing rare nodes rather than rare observations (data points), and focusing on uncertain nodes for each model output distribution during training, we observe improvements in recall by up to a factor of five on benchmark datasets, along with statistically significant gains in $F_{1}$ score. We also show our approach aids convolutional networks on challenging tasks, as in situations with suboptimal encoders or limited data.

cs.LG↗

arXiv2026

Variance-Gated Ensembles: An Epistemic-Aware Framework for Uncertainty Estimation

Machine learning applications require fast and reliable per-sample uncertainty estimation. A common approach is to use predictive distributions from Bayesian or approximation methods and additively decompose uncertainty into aleatoric (i.e., data-related) and epistemic (i.e., model-related) components. However, additive decomposition has recently been questioned, with evidence that it breaks down when using finite-ensemble sampling and/or mismatched predictive distributions. This paper introduces Variance-Gated Ensembles (VGE), an intuitive, differentiable framework that injects epistemic sensitivity via a signal-to-noise gate computed from ensemble statistics. VGE provides: (i) a Variance-Gated Margin Uncertainty (VGMU) score that couples decision margins with ensemble predictive variance; and (ii) a Variance-Gated Normalization (VGN) layer that generalizes the variance-gated uncertainty mechanism to training via per-class, learnable normalization of ensemble member probabilities. We derive closed-form vector-Jacobian products enabling end-to-end training through ensemble sample mean and variance. VGE matches or exceeds state-of-the-art information-theoretic baselines while remaining computationally efficient. As a result, VGE provides a practical and scalable approach to epistemic-aware uncertainty estimation in ensemble models.

cs.LG↗

arXiv2025

Uncertainty Estimation using Variance-Gated Distributions

Evaluation of per-sample uncertainty quantification from neural networks is essential for decision-making involving high-risk applications. A common approach is to use the predictive distribution from Bayesian or approximation models and decompose the corresponding predictive uncertainty into epistemic (model-related) and aleatoric (data-related) components. However, additive decomposition has recently been questioned. In this work, we propose an intuitive framework for uncertainty estimation and decomposition based on the signal-to-noise ratio of class probability distributions across different model predictions. We introduce a variance-gated measure that scales predictions by a confidence factor derived from ensembles. We use this measure to discuss the existence of a collapse in the diversity of committee machines.

cs.LG↗

arXiv2025

Platelet enumeration in dense aggregates

Identifying and counting blood components such as red blood cells, various types of white blood cells, and platelets is a critical task for healthcare practitioners. Deep learning approaches, particularly convolutional neural networks (CNNs) using supervised learning strategies, have shown considerable success for such tasks. However, CNN based architectures such as U-Net, often struggles to accurately identify platelets due to their sizes and high variability of features. To address these challenges, researchers have commonly employed strategies such as class weighted loss functions, which have demonstrated some success. However, this does not address the more significant challenge of platelet variability in size and tendency to form aggregates and associations with other blood components. In this study, we explored an alternative approach by investigating the role of convolutional kernels in mitigating these issues. We also assigned separate classes to singular platelets and platelet aggregates and performed semantic segmentation using various U-Net architectures for identifying platelets. We then evaluated and compared two common methods (pixel area method and connected component analysis) for counting platelets and proposed an alternative approach specialized for single platelets and platelet aggregates. Our experiments provided results that showed significant improvements in the identification of platelets, highlighting the importance of optimizing convolutional operations and class designations. We show that the common practice of pixel area-based counting often over estimate platelet counts, whereas the proposed method presented in this work offers significant improvements. We discuss in detail about these methods from segmentation masks.

eess.IV↗

arXiv2025

Masked strategies for images with small objects

The hematology analytics used for detection and classification of small blood components is a significant challenge. In particular, when objects exists as small pixel-sized entities in a large context of similar objects. Deep learning approaches using supervised models with pre-trained weights, such as residual networks and vision transformers have demonstrated success for many applications. Unfortunately, when applied to images outside the domain of learned representations, these methods often result with less than acceptable performance. A strategy to overcome this can be achieved by using self-supervised models, where representations are learned and weights are then applied for downstream applications. Recently, masked autoencoders have proven to be effective to obtain representations that captures global context information. By masking regions of an image and having the model learn to reconstruct both the masked and non-masked regions, weights can be used for various applications. However, if the sizes of the objects in images are less than the size of the mask, the global context information is lost, making it almost impossible to reconstruct the image. In this study, we investigated the effect of mask ratios and patch sizes for blood components using a MAE to obtain learned ViT encoder representations. We then applied the encoder weights to train a U-Net Transformer for semantic segmentation to obtain both local and global contextual information. Our experimental results demonstrates that both smaller mask ratios and patch sizes improve the reconstruction of images using a MAE. We also show the results of semantic segmentation with and without pre-trained weights, where smaller-sized blood components benefited with pre-training. Overall, our proposed method offers an efficient and effective strategy for the segmentation and classification of small objects.

cs.CV↗

arXiv2025

Last-layer committee machines for uncertainty estimations of benthic imagery

Automating the annotation of benthic imagery (i.e., images of the seafloor and its associated organisms, habitats, and geological features) is critical for monitoring rapidly changing ocean ecosystems. Deep learning approaches have succeeded in this purpose; however, consistent annotation remains challenging due to ambiguous seafloor images, potential inter-user annotation disagreements, and out-of-distribution samples. Marine scientists implementing deep learning models often obtain predictions based on one-hot representations trained using a cross-entropy loss objective with softmax normalization, resulting with a single set of model parameters. While efficient, this approach may lead to overconfident predictions for context-challenging datasets, raising reliability concerns that present risks for downstream tasks such as benthic habitat mapping and marine spatial planning. In this study, we investigated classification uncertainty as a tool to improve the labeling of benthic habitat imagery. We developed a framework for two challenging sub-datasets of the recently publicly available BenthicNet dataset using Bayesian neural networks, Monte Carlo dropout inference sampling, and a proposed single last-layer committee machine. This approach resulted with a > 95% reduction of network parameters to obtain per-sample uncertainties while obtaining near-identical performance compared to computationally more expensive strategies such as Bayesian neural networks, Monte Carlo dropout, and deep ensembles. The method proposed in this research provides a strategy for obtaining prioritized lists of uncertain samples for human-in-the-loop interventions to identify ambiguous, mislabeled, out-of-distribution, and/or difficult images for enhancing existing annotation tools for benthic mapping and other applications.

q-bio.QM↗

arXiv2024

Label-free Monitoring of Self-Supervised Learning Progress

Self-supervised learning (SSL) is an effective method for exploiting unlabelled data to learn a high-level embedding space that can be used for various downstream tasks. However, existing methods to monitor the quality of the encoder -- either during training for one model or to compare several trained models -- still rely on access to annotated data. When SSL methodologies are applied to new data domains, a sufficiently large labelled dataset may not always be available. In this study, we propose several evaluation metrics which can be applied on the embeddings of unlabelled data and investigate their viability by comparing them to linear probe accuracy (a common metric which utilizes an annotated dataset). In particular, we apply $k$-means clustering and measure the clustering quality with the silhouette score and clustering agreement. We also measure the entropy of the embedding distribution. We find that while the clusters did correspond better to the ground truth annotations as training of the network progressed, label-free clustering metrics correlated with the linear probe accuracy only when training with SSL methods SimCLR and MoCo-v2, but not with SimSiam. Additionally, although entropy did not always have strong correlations with LP accuracy, this appears to be due to instability arising from early training, with the metric stabilizing and becoming more reliable at later stages of learning. Furthermore, while entropy generally decreases as learning progresses, this trend reverses for SimSiam. More research is required to establish the cause for this unexpected behaviour. Lastly, we find that while clustering based approaches are likely only viable for same-architecture comparisons, entropy may be architecture-independent.

cs.LG↗

arXiv2024

Hierarchical Multi-Label Classification with Missing Information for Benthic Habitat Imagery

In this work, we apply state-of-the-art self-supervised learning techniques on a large dataset of seafloor imagery, \textit{BenthicNet}, and study their performance for a complex hierarchical multi-label (HML) classification downstream task. In particular, we demonstrate the capacity to conduct HML training in scenarios where there exist multiple levels of missing annotation information, an important scenario for handling heterogeneous real-world data collected by multiple research groups with differing data collection protocols. We find that, when using smaller one-hot image label datasets typical of local or regional scale benthic science projects, models pre-trained with self-supervision on a larger collection of in-domain benthic data outperform models pre-trained on ImageNet. In the HML setting, we find the model can attain a deeper and more precise classification if it is pre-trained with self-supervision on in-domain data. We hope this work can establish a benchmark for future models in the field of automated underwater image annotation tasks and can guide work in other domains with hierarchical annotations of mixed resolution.

cs.CV↗

arXiv2024

An Information-Geometric Formulation of Pattern Separation and Evaluation of Existing Indices

Pattern separation is a computational process by which dissimilar neural patterns are generated from similar input patterns. We present an information-geometric formulation of pattern separation, where a pattern separator is modelled as a family of statistical distributions on a manifold. Such a manifold maps an input (i.e. coordinates) to a probability distribution that generates firing patterns. Pattern separation occurs when small coordinate changes result in large distances between samples from the corresponding distributions. Under this formulation, we implement a two-neuron system whose probability law forms a 3-dimensional manifold with mutually orthogonal coordinates representing the neurons' marginal and correlational firing rates. We use this highly controlled system to examine the behaviour of spike train similarity indices commonly used in pattern separation research. We found that all indices (except scaling factor) were sensitive to relative differences in marginal firing rates, but no index adequately captured differences in spike trains that resulted from altering the correlation in activity between the two neurons. That is, existing pattern separation metrics appear (A) sensitive to patterns that are encoded by different neurons, but (B) insensitive to patterns that differ only in relative spike timing (e.g. synchrony between neurons in the ensemble).

q-bio.NC↗

arXiv2024

A Generalized Transformer-based Radio Link Failure Prediction Framework in 5G RANs

Radio link failure (RLF) prediction system in Radio Access Networks (RANs) is critical for ensuring seamless communication and meeting the stringent requirements of high data rates, low latency, and improved reliability in 5G networks. However, weather conditions such as precipitation, humidity, temperature, and wind impact these communication links. Usually, historical radio link Key Performance Indicators (KPIs) and their surrounding weather station observations are utilized for building learning-based RLF prediction models. However, such models must be capable of learning the spatial weather context in a dynamic RAN and effectively encoding time series KPIs with the weather observation data. Existing works fail to incorporate both of these essential design aspects of the prediction models. This paper fills the gap by proposing GenTrap, a novel RLF prediction framework that introduces a graph neural network (GNN)-based learnable weather effect aggregation module and employs state-of-the-art time series transformer as the temporal feature extractor for radio link failure prediction. The proposed aggregation method of GenTrap can be integrated into any existing prediction model to achieve better performance and generalizability. We evaluate GenTrap on two real-world datasets (rural and urban) with 2.6 million KPI data points and show that GenTrap offers a significantly higher F1-score (0.93 for rural and 0.79 for urban) compared to its counterparts while possessing generalization capability.

cs.AI↗

arXiv2024

BenthicNet: A global compilation of seafloor images for deep learning applications

Advances in underwater imaging enable collection of extensive seafloor image datasets necessary for monitoring important benthic ecosystems. The ability to collect seafloor imagery has outpaced our capacity to analyze it, hindering mobilization of this crucial environmental information. Machine learning approaches provide opportunities to increase the efficiency with which seafloor imagery is analyzed, yet large and consistent datasets to support development of such approaches are scarce. Here we present BenthicNet: a global compilation of seafloor imagery designed to support the training and evaluation of large-scale image recognition models. An initial set of over 11.4 million images was collected and curated to represent a diversity of seafloor environments using a representative subset of 1.3 million images. These are accompanied by 3.1 million annotations translated to the CATAMI scheme, which span 190,000 of the images. A large deep learning model was trained on this compilation and preliminary results suggest it has utility for automating large and small-scale image analysis tasks. The compilation and model are made openly available for reuse at https://doi.org/10.20383/103.0614.

cs.CV↗

arXiv2021

LogAvgExp Provides a Principled and Performant Global Pooling Operator

We seek to improve the pooling operation in neural networks, by applying a more theoretically justified operator. We demonstrate that LogSumExp provides a natural OR operator for logits. When one corrects for the number of elements inside the pooling operator, this becomes $\text{LogAvgExp} := \log(\text{mean}(\exp(x)))$. By introducing a single temperature parameter, LogAvgExp smoothly transitions from the max of its operands to the mean (found at the limiting cases $t \to 0^+$ and $t \to +\infty$). We experimentally tested LogAvgExp, both with and without a learnable temperature parameter, in a variety of deep neural network architectures for computer vision.

cs.LG↗

arXiv2021

Logical Activation Functions: Logit-space equivalents of Probabilistic Boolean Operators

The choice of activation functions and their motivation is a long-standing issue within the neural network community. Neuronal representations within artificial neural networks are commonly understood as logits, representing the log-odds score of presence of features within the stimulus. We derive logit-space operators equivalent to probabilistic Boolean logic-gates AND, OR, and XNOR for independent probabilities. Such theories are important to formalize more complex dendritic operations in real neurons, and these operations can be used as activation functions within a neural network, introducing probabilistic Boolean-logic as the core operation of the neural network. Since these functions involve taking multiple exponents and logarithms, they are computationally expensive and not well suited to be directly used within neural networks. Consequently, we construct efficient approximations named $\text{AND}_\text{AIL}$ (the AND operator Approximate for Independent Logits), $\text{OR}_\text{AIL}$, and $\text{XNOR}_\text{AIL}$, which utilize only comparison and addition operations, have well-behaved gradients, and can be deployed as activation functions in neural networks. Like MaxOut, $\text{AND}_\text{AIL}$ and $\text{OR}_\text{AIL}$ are generalizations of ReLU to two-dimensions. While our primary aim is to formalize dendritic computations within a logit-space probabilistic-Boolean framework, we deploy these new activation functions, both in isolation and in conjunction to demonstrate their effectiveness on a variety of tasks including image classification, transfer learning, abstract reasoning, and compositional zero-shot learning.

cs.LG↗

arXiv2020

Multiplicative Decomposition of Heterogeneity in Mixtures of Continuous Distributions

A system's heterogeneity (\textit{diversity}) is the effective size of its event space, and can be quantified using the R\'enyi family of indices (also known as Hill numbers in ecology or Hannah-Kay indices in economics), which are indexed by an elasticity parameter $q \geq 0$. Under these indices, the heterogeneity of a composite system (the $\gamma$-heterogeneity) is decomposable into heterogeneity arising from variation \textit{within} and \textit{between} component subsystems (the $\alpha$- and $\beta$-heterogeneity, respectively). Since the average heterogeneity of a component subsystem should not be greater than that of the pooled system, we require that $\gamma \geq \alpha$. There exists a multiplicative decomposition for R\'enyi heterogeneity of composite systems with discrete event spaces, but less attention has been paid to decomposition in the continuous setting. We therefore describe multiplicative decomposition of the R\'enyi heterogeneity for continuous mixture distributions under parametric and non-parametric pooling assumptions. Under non-parametric pooling, the $\gamma$-heterogeneity must often be estimated numerically, but the multiplicative decomposition holds such that $\gamma \geq \alpha$ for $q > 0$. Conversely, under parametric pooling, $\gamma$-heterogeneity can be computed efficiently in closed-form, but the $\gamma \geq \alpha$ condition holds reliably only at $q=1$. Our findings will further contribute to heterogeneity measurement in continuous systems.

q-bio.QM↗

arXiv2019

Representational R\'enyi heterogeneity

A discrete system's heterogeneity is measured by the R\'enyi heterogeneity family of indices (also known as Hill numbers or Hannah--Kay indices), whose units are {the numbers equivalent}. Unfortunately, numbers equivalent heterogeneity measures for non-categorical data require {a priori} (A) categorical partitioning and (B) pairwise distance measurement on the observable data space, thereby precluding application to problems with ill-defined categories or where semantically relevant features must be learned as abstractions from some data. We thus introduce representational R\'enyi heterogeneity (RRH), which transforms an observable domain onto a latent space upon which the R\'enyi heterogeneity is both tractable and semantically relevant. This method requires neither {a priori} binning nor definition of a distance function on the observable space. We show that RRH can generalize existing biodiversity and economic equality indices. Compared with existing indices on a beta-mixture distribution, we show that RRH responds more appropriately to changes in mixture component separation and weighting. Finally, we demonstrate the measurement of RRH in a set of natural images, with respect to abstract representations learned by a deep neural network. The RRH approach will further enable heterogeneity measurement in disciplines whose data do not easily conform to the assumptions of existing indices.

stat.ML↗

arXiv2019

Skin cancer detection based on deep learning and entropy to detect outlier samples

We describe our methods that achieved the 3rd and 4th places in tasks 1 and 2, respectively, at ISIC challenge 2019. The goal of this challenge is to provide the diagnostic for skin cancer using images and meta-data. There are nine classes in the dataset, nonetheless, one of them is an outlier and is not present on it. To tackle the challenge, we apply an ensemble of classifiers, which has 13 convolutional neural networks (CNN), we develop two approaches to handle the outlier class and we propose a straightforward method to use the meta-data along with the images. Throughout this report, we detail each methodology and parameters to make it easy to replicate our work. The results obtained are in accordance with the previous challenges and the approaches to detect the outlier class and to address the meta-data seem to be work properly.

cs.LG↗