SearcharxivSearch

arXiv subjects

Martin Atzmueller

Publications and source records attributed to Martin Atzmueller.

At least 37 records · Page 2Linked to original sources

Comprehensive Evaluation of Prototype Neural Networks

Prototype models are an important method for explainable artificial intelligence (XAI) and interpretable machine learning. In this paper, we perform an in-depth analysis of a set of prominent prototype models including ProtoPNet, ProtoPool and PIPNet. For their assessment, we apply a comprehensive set of metrics. In addition to applying standard metrics from literature, we propose several new metrics to further complement the analysis of model interpretability. In our experimentation, we apply the set of prototype models on a diverse set of datasets including fine-grained classification, Non-IID settings and multi-label classification to further contrast the performance. Furthermore, we also provide our code as an open-source library (https://github.com/uos-sis/quanproto), which facilitates simple application of the metrics itself, as well as extensibility -- providing the option for easily adding new metrics and models.

cs.LG

PHEATPRUNER: Interpretable Data-centric Feature Selection for Multivariate Time Series Classification through Persistent Homology

Balancing performance and interpretability in multivariate time series classification is a significant challenge due to data complexity and high dimensionality. This paper introduces PHeatPruner, a method integrating persistent homology and sheaf theory to address these challenges. Persistent homology facilitates the pruning of up to 45% of the applied variables while maintaining or enhancing the accuracy of models such as Random Forest, CatBoost, XGBoost, and LightGBM, all without depending on posterior probabilities or supervised optimization algorithms. Concurrently, sheaf theory contributes explanatory vectors that provide deeper insights into the data's structural nuances. The approach was validated using the UEA Archive and a mastitis detection dataset for dairy cows. The results demonstrate that PHeatPruner effectively preserves model accuracy. Furthermore, our results highlight PHeatPruner's key features, i.e. simplifying complex data and offering actionable insights without increasing processing time or complexity. This method bridges the gap between complexity reduction and interpretability, suggesting promising applications in various fields.

cs.LG

Saliency Maps are Ambiguous: Analysis of Logical Relations on First and Second Order Attributions

Recent work uncovered potential flaws in \eg attribution or heatmap based saliency methods. A typical flaw is a confirmations bias, where the scores are compared to human expectation. Since measuring the quality of saliency methods is hard due to missing ground truth model reasoning, finding general limitations is also hard. This is further complicated, because masking-based evaluation on complex data can easily introduce a bias, as most methods cannot fully ignore inputs. In this work, we extend our previous analysis on the logical dataset framework ANDOR, where we showed that all analysed saliency methods fail to grasp all needed classification information for all possible scenarios. Specifically, this paper extends our previous work using analysis on more datasets, in order to better understand in which scenarios the saliency methods fail. Further, we apply the Global Coherence Representation as an additional evaluation method in order to enable actual input omission.

cs.LG

Saliency Methods are Encoders: Analysing Logical Relations Towards Interpretation

With their increase in performance, neural network architectures also become more complex, necessitating explainability. Therefore, many new and improved methods are currently emerging, which often generate so-called saliency maps in order to improve interpretability. Those methods are often evaluated by visual expectations, yet this typically leads towards a confirmation bias. Due to a lack of a general metric for explanation quality, non-accessible ground truth data about the model's reasoning and the large amount of involved assumptions, multiple works claim to find flaws in those methods. However, this often leads to unfair comparison metrics. Additionally, the complexity of most datasets (mostly images or text) is often so high, that approximating all possible explanations is not feasible. For those reasons, this paper introduces a test for saliency map evaluation: proposing controlled experiments based on all possible model reasonings over multiple simple logical datasets. Using the contained logical relationships, we aim to understand how different saliency methods treat information in different class discriminative scenarios (e.g. via complementary and redundant information). By introducing multiple new metrics, we analyse propositional logical patterns towards a non-informative attribution score baseline to find deviations of typical expectations. Our results show that saliency methods can encode classification relevant information into the ordering of saliency scores.

cs.LG

Knowledge-Augmented Explainable and Interpretable Learning for Anomaly Detection and Diagnosis

Knowledge-augmented learning enables the combination of knowledge-based and data-driven approaches. For anomaly detection and diagnosis, understandability is typically an important factor, especially in high-risk areas. Therefore, explainability and interpretability are also major criteria in such contexts. This chapter focuses on knowledge-augmented explainable and interpretable learning to enhance understandability, transparency and ultimately computational sensemaking. We exemplify different approaches and methods in the domains of anomaly detection and diagnosis - from comparatively simple interpretable methods towards more advanced neuro-symbolic approaches.

cs.LG

Online Knowledge Integration for 3D Semantic Mapping: A Survey

Semantic mapping is a key component of robots operating in and interacting with objects in structured environments. Traditionally, geometric and knowledge representations within a semantic map have only been loosely integrated. However, recent advances in deep learning now allow full integration of prior knowledge, represented as knowledge graphs or language concepts, into sensor data processing and semantic mapping pipelines. Semantic scene graphs and language models enable modern semantic mapping approaches to incorporate graph-based prior knowledge or to leverage the rich information in human language both during and after the mapping process. This has sparked substantial advances in semantic mapping, leading to previously impossible novel applications. This survey reviews these recent developments comprehensively, with a focus on online integration of knowledge into semantic mapping. We specifically focus on methods using semantic scene graphs for integrating symbolic prior knowledge and language models for respective capture of implicit common-sense knowledge and natural language concepts

cs.RO

Finding One's Bearings in the Hyperparameter Landscape of a Wide-Kernel Convolutional Fault Detector

State-of-the-art algorithms are reported to be almost perfect at distinguishing the vibrations arising from healthy and damaged machine bearings, according to benchmark datasets at least. However, what about their application to new data? In this paper, we confirm that neural networks for bearing fault detection can be crippled by incorrect hyperparameterisation, and also that the correct hyperparameter settings can change when transitioning to new data. The paper combines multiple methods to explain the behaviour of the hyperparameters of a wide-kernel convolutional neural network and how to set them. Since guidance already exists for generic hyperparameters like minibatch size, we focus on how to set architecture-specific hyperparameters such as the width of the convolutional kernels, a topic which might otherwise be obscure. We reflect different data properties by fusing information from seven different benchmark datasets, and our results show that the kernel size in the first layer in particular is sensitive to changes in the data. Looking deeper, we use manipulated copies of one dataset in an attempt to spot why the kernel size sometimes needs to change. The relevance of sampling rate is studied by using different levels of resampling, and spectral content is studied by increasingly filtering out high frequencies. We find that, contrary to speculation in earlier work, high-frequency noise is not the main reason why a wide kernel is preferable to a narrow kernel. Finally, we conclude by stating clear guidance on how to set the hyperparameters of our neural network architecture to work effectively on new data.

cs.LG

An Efficient Model-Agnostic Approach for Uncertainty Estimation in Data-Restricted Pedometric Applications

This paper introduces a model-agnostic approach designed to enhance uncertainty estimation in the predictive modeling of soil properties, a crucial factor for advancing pedometrics and the practice of digital soil mapping. For addressing the typical challenge of data scarcity in soil studies, we present an improved technique for uncertainty estimation. This method is based on the transformation of regression tasks into classification problems, which not only allows for the production of reliable uncertainty estimates but also enables the application of established machine learning algorithms with competitive performance that have not yet been utilized in pedometrics. Empirical results from datasets collected from two German agricultural fields showcase the practical application of the proposed methodology. Our results and findings suggest that the proposed approach has the potential to provide better uncertainty estimation than the models commonly used in pedometrics.

cs.LG

An Empirical Study on Improving SimCLR's Nonlinear Projection Head using Pretrained Autoencoder Embeddings

This paper focuses on improving the effectiveness of the standard 2-layer MLP projection head featured in the SimCLR framework through the use of pretrained autoencoder embeddings. Given a contrastive learning task with a largely unlabeled image classification dataset, we first train a shallow autoencoder architecture and extract its compressed representations contained in the encoder's embedding layer. After freezing the weights within this pretrained layer, we use it as a drop-in replacement for the input layer of SimCLR's default projector. Additionally, we also apply further architectural changes to the projector by decreasing its width and changing its activation function. The different projection heads are then used to contrastively train and evaluate a feature extractor following the SimCLR protocol. Our experiments indicate that using a pretrained autoencoder embedding in the projector can not only increase classification accuracy by up to 2.9% or 1.7% on average, but can also significantly decrease the dimensionality of the projection space. Our results also suggest, that using the sigmoid and tanh activation functions within the projector can outperform ReLU in terms of peak and average classification accuracy. All experiments involving our pretrained projectors are conducted with frozen embeddings, since our test results indicate an advantage compared to using their non-frozen counterparts.

cs.LG

Simplicial complexes in network intrusion profiling

For studying intrusion detection data we consider data points referring to individual IP addresses and their connections: We build networks associated with those data points, such that vertices in a graph are associated via the respective IP addresses, with the key property that attacked data points are part of the structure of the network. More precisely, we propose a novel approach using simplicial complexes to model the desired network and the respective intrusions in terms of simplicial attributes thus generalizing previous graph-based approaches. Adapted network centrality measures related to simplicial complexes yield so-called patterns associated to vertices, which themselves contain a set of features. These are then used to describe the attacked or the attacker vertices, respectively. Comparing this new strategy with classical concepts demonstrates the advantages of the presented approach using simplicial features for detecting and characterizing intrusions.

math.AC

Extracting Interpretable Local and Global Representations from Attention on Time Series

This paper targets two transformer attention based interpretability methods working with local abstraction and global representation, in the context of time series data. We distinguish local and global contexts, and provide a comprehensive framework for both general interpretation options. We discuss their specific instantiation via different methods in detail, also outlining their respective computational implementation and abstraction variants. Furthermore, we provide extensive experimentation demonstrating the efficacy of the presented approaches. In particular, we perform our experiments using a selection of univariate datasets from the UCR UEA time series repository where we both assess the performance of the proposed approaches, as well as their impact on explainability and interpretability/complexity. Here, with an extensive analysis of hyperparameters, the presented approaches demonstrate an significant improvement in interpretability/complexity, while capturing many core decisions of and maintaining a similar performance to the baseline model. Finally, we draw general conclusions outlining and guiding the application of the presented methods.

cs.LG

Graph Neural Networks for Multivariate Time Series Regression with Application to Seismic Data

Machine learning, with its advances in deep learning has shown great potential in analyzing time series. In many scenarios, however, additional information that can potentially improve the predictions is available. This is crucial for data that arise from e.g., sensor networks that contain information about sensor locations. Then, such spatial information can be exploited by modeling it via graph structures, along with the sequential (time series) information. Recent advances in adapting deep learning to graphs have shown potential in various tasks. However, these methods have not been adapted for time series tasks to a great extent. Most attempts have essentially consolidated around time series forecasting with small sequence lengths. Generally, these architectures are not well suited for regression or classification tasks where the value to be predicted is not strictly depending on the most recent values, but rather on the whole length of the time series. We propose TISER-GCN, a novel graph neural network architecture for processing, in particular, these long time series in a multivariate regression task. Our proposed model is tested on two seismic datasets containing earthquake waveforms, where the goal is to predict maximum intensity measurements of ground shaking at each seismic station. Our findings demonstrate promising results of our approach -- with an average MSE reduction of 16.3% - compared to the best performing baselines. In addition, our approach matches the baseline scores by needing only half the input size. The results are discussed in depth with an additional ablation study.

cs.LG

Local Exceptionality Detection in Time Series Using Subgroup Discovery

In this paper, we present a novel approach for local exceptionality detection on time series data. This method provides the ability to discover interpretable patterns in the data, which can be used to understand and predict the progression of a time series. This being an exploratory approach, the results can be used to generate hypotheses about the relationships between the variables describing a specific process and its dynamics. We detail our approach in a concrete instantiation and exemplary implementation, specifically in the field of teamwork research. Using a real-world dataset of team interactions we include results from an example data analytics application of our proposed approach, showcase novel analysis options, and discuss possible implications of the results from the perspective of teamwork research.

cs.LG

A Computational Framework for Modeling Complex Sensor Network Data Using Graph Signal Processing and Graph Neural Networks in Structural Health Monitoring

Complex networks lend themselves to the modeling of multidimensional data, such as relational and/or temporal data. In particular, when such complex data and their inherent relationships need to be formalized, complex network modeling and its resulting graph representations enable a wide range of powerful options. In this paper, we target this - connected to specific machine learning approaches on graphs for structural health monitoring on an analysis and predictive (maintenance) perspective. Specifically, we present a framework based on Complex Network Modeling, integrating Graph Signal Processing (GSP) and Graph Neural Network (GNN) approaches. We demonstrate this framework in our targeted application domain of Structural Health Monitoring (SHM). In particular, we focus on a prominent real-world structural health monitoring use case, i.e., modeling and analyzing sensor data (strain, vibration) of a large bridge in the Netherlands. In our experiments, we show that GSP enables the identification of the most important sensors, for which we investigate a set of search and optimization approaches. Furthermore, GSP enables the detection of specific graph signal patterns (mode shapes), capturing physical functional properties of the sensors in the applied complex network. In addition, we show the efficacy of applying GNNs for strain prediction on this kind of data.

cs.LG

Social Analytics of Team Interaction using Dynamic Complexity Heat Maps and Network Visualizations

Given the increasing complexity of many sociotechnical work domains, effective teamwork has become increasingly crucial. While there is evidence that face-to-face communication contributes to effective teamwork, methods for understanding the time-varying nature and structure of team communication are limited. In this work, we combine sensor-based social analytics of Sociometric badges (Rhythm Badge) with two visualization techniques (Dynamic Complexity Heat Maps and Network Visualizations) to advance an intuitive way of understanding the dynamics of team interaction. To demonstrate the utility of our approach, we provide a case study that examines one team's interaction for a Lost at Sea simulation. We were able to recover transitions in the task and team interaction as well as uncover structural changes in team member energy and engagement, which we visualize using networks. Taken together, this work represents an important first step at optimizing team effectiveness by identifying critical transitions/events in team communication and interaction patterns.

cs.SI

Towards Generating Explanations for ASP-Based Link Analysis using Declarative Program Transformations

The explication and the generation of explanations are prominent topics in artificial intelligence and data science, in order to make methods and systems more transparent and understandable for humans. This paper investigates the problem of link analysis, specifically link prediction and anomalous link discovery in social networks using the declarative method of Answer set programming (ASP). Applying ASP for link prediction provides a powerful declarative approach, e.g., for incorporating domain knowledge for explicative prediction. In this context, we propose a novel method for generating explanations - as offline justifications - using declarative program transformations. The method itself is purely based on syntactic transformations of declarative programs, e.g., in an ASP formalism, using rule instrumentation. We demonstrate the efficacy of the proposed approach, exemplifying it in an application on link analysis in social networks, also including domain knowledge.

cs.AI

Feature-Set-Engineering for Detecting Freezing of Gait in Parkinson's Disease using Deep Recurrent Neural Networks

Freezing of gait (FoG) is a common gait disability in Parkinson's disease, that usually appears in its advanced stage. Freeze episodes are associated with falls, injuries, and psychological consequences, negatively affecting the patients' quality of life. For detecting FoG episodes automatically, a highly accurate detection method is necessary. This paper presents an approach for detecting FoG episodes utilizing a deep recurrent neural network (RNN) on 3D-accelerometer measurements. We investigate suitable features and feature combinations extracted from the sensors' time series data. Specifically, for detecting FoG episodes, we apply a deep RNN with Long Short-Term Memory cells. In our experiments, we perform both user dependent and user independent experiments, to detect freeze episodes. Our experimental results show that the frequency domain features extracted from the trunk sensor are the most informative feature group in the subject independent method, achieving an average AUC score of 93%, Specificity of 90% and Sensitivity of 81%. Moreover, frequency and statistical features of all the sensors are identified as the best single input for the subject dependent method, achieving an average AUC score of 97%, Specificity of 96% and Sensitivity of 87%. Overall, in a comparison to state-of-the-art approaches from literature as baseline methods, our proposed approach outperforms these significantly.

eess.SP

Finding Maximal Non-Redundant Association Rules in Tennis Data

The concept of association rules is well--known in data mining. But often redundancy and subsumption are not considered, and standard approaches produce thousands or even millions of resulting association rules. Without further information or post--mining approaches, this huge number of rules is typically useless for the domain specialist -- which is an instance of the infamous pattern explosion problem. In this work, we present a new definition of redundancy and subsumption based on the confidence and the support of the rules and propose post-- mining to prune a set of association rules. In a case study, we apply our method to association rules mined from spatio--temporal data. The data represent the trajectories of the ball in tennis matches -- more precisely, the points/times the tennis ball hits the ground. The goal is to analyze the strategies of the players and to try to improve their performance by looking at the resulting association rules. The proposed approach is general, and can also be applied to other spatio--temporal data with a similar structure.

cs.DB