arXiv subjects

Sofie Van Hoecke

Publications and source records attributed to Sofie Van Hoecke.

25 records · Page 2Linked to original sources

arXiv2021

tsflex: flexible time series processing & feature extraction

Time series processing and feature extraction are crucial and time-intensive steps in conventional machine learning pipelines. Existing packages are limited in their applicability, as they cannot cope with irregularly-sampled or asynchronous data and make strong assumptions about the data format. Moreover, these packages do not focus on execution speed and memory efficiency, resulting in considerable overhead. We present $\texttt{tsflex}$, a Python toolkit for time series processing and feature extraction, that focuses on performance and flexibility, enabling broad applicability. This toolkit leverages window-stride arguments of the same data type as the sequence-index, and maintains the sequence-index through all operations. $\texttt{tsflex}$ is flexible as it supports (1) multivariate time series, (2) multiple window-stride configurations, and (3) integrates with processing and feature functions from other packages, while (4) making no assumptions about the data sampling regularity, series alignment, and data type. Other functionalities include multiprocessing, detailed execution logging, chunking sequences, and serialization. Benchmarks show that $\texttt{tsflex}$ is faster and more memory-efficient compared to similar packages, while being more permissive and flexible in its utilization.

cs.LG↗

arXiv2021

Neural Network Augmented Physics Models for Systems with Partially Unknown Dynamics: Application to Slider-Crank Mechanism

Dynamic models of mechatronic systems are abundantly used in the context of motion control and design of complex servo applications. In practice, these systems are often plagued by unknown interactions, which make the physics-based relations of the system dynamics only partially known. This paper presents a neural network augmented physics (NNAP) model as a combination of physics-inspired and neural layers. The neural layers are inserted in the model to compensate for the unmodeled interactions, without requiring direct measurements of these unknown phenomena. In contrast to traditional approaches, both the neural network and physical parameters are simultaneously optimized, solely by using state and control input measurements. The methodology is applied on experimental data of a slider-crank setup for which the state dependent load interactions are unknown. The NNAP model proves to be a stable and accurate modeling formalism for dynamic systems that ab initio can only be partially described by physical laws. Moreover, the results show that a recurrent implementation of the NNAP model enables improved robustness and accuracy of the system state predictions, compared to its feedforward counterpart. Besides capturing the system dynamics, the NNAP model provides a means to gain new insights by extracting the neural network from the converged NNAP model. In this way, we discovered accurate representations of the unknown spring force interaction and friction phenomena acting on the slider mechanism.

eess.SY↗

arXiv2020

Overly Optimistic Prediction Results on Imbalanced Data: a Case Study of Flaws and Benefits when Applying Over-sampling

Information extracted from electrohysterography recordings could potentially prove to be an interesting additional source of information to estimate the risk on preterm birth. Recently, a large number of studies have reported near-perfect results to distinguish between recordings of patients that will deliver term or preterm using a public resource, called the Term/Preterm Electrohysterogram database. However, we argue that these results are overly optimistic due to a methodological flaw being made. In this work, we focus on one specific type of methodological flaw: applying over-sampling before partitioning the data into mutually exclusive training and testing sets. We show how this causes the results to be biased using two artificial datasets and reproduce results of studies in which this flaw was identified. Moreover, we evaluate the actual impact of over-sampling on predictive performance, when applied prior to data partitioning, using the same methodologies of related studies, to provide a realistic view of these methodologies' generalization capabilities. We make our research reproducible by providing all the code under an open license.

eess.SP↗

arXiv2019

Web Applicable Computer-aided Diagnosis of Glaucoma Using Deep Learning

Glaucoma is a major eye disease, leading to vision loss in the absence of proper medical treatment. Current diagnosis of glaucoma is performed by ophthalmologists who are often analyzing several types of medical images generated by different types of medical equipment. Capturing and analyzing these medical images is labor-intensive and expensive. In this paper, we present a novel computational approach towards glaucoma diagnosis and localization, only making use of eye fundus images that are analyzed by state-of-the-art deep learning techniques. Specifically, our approach leverages Convolutional Neural Networks (CNNs) and Gradient-weighted Class Activation Mapping (Grad-CAM) for glaucoma diagnosis and localization, respectively. Quantitative and qualitative results, as obtained for a small-sized dataset with no segmentation ground truth, demonstrate that the proposed approach is promising, for instance achieving an accuracy of 0.91$\pm0.02$ and an ROC-AUC score of 0.94 for the diagnosis task. Furthermore, we present a publicly available prototype web application that integrates our predictive model, with the goal of making effective glaucoma diagnosis available to a wide audience.

cs.CV↗

arXiv2016

GENESIM: genetic extraction of a single, interpretable model

Models obtained by decision tree induction techniques excel in being interpretable.However, they can be prone to overfitting, which results in a low predictive performance. Ensemble techniques are able to achieve a higher accuracy. However, this comes at a cost of losing interpretability of the resulting model. This makes ensemble techniques impractical in applications where decision support, instead of decision making, is crucial. To bridge this gap, we present the GENESIM algorithm that transforms an ensemble of decision trees to a single decision tree with an enhanced predictive performance by using a genetic algorithm. We compared GENESIM to prevalent decision tree induction and ensemble techniques using twelve publicly available data sets. The results show that GENESIM achieves a better predictive performance on most of these data sets than decision tree induction techniques and a predictive performance in the same order of magnitude as the ensemble techniques. Moreover, the resulting model of GENESIM has a very low complexity, making it very interpretable, in contrast to ensemble techniques.

stat.ML↗

arXiv2015

The Pragmatic Proof: Hypermedia API Composition and Execution

Machine clients are increasingly making use of the Web to perform tasks. While Web services traditionally mimic remote procedure calling interfaces, a new generation of so-called hypermedia APIs works through hyperlinks and forms, in a way similar to how people browse the Web. This means that existing composition techniques, which determine a procedural plan upfront, are not sufficient to consume hypermedia APIs, which need to be navigated at runtime. Clients instead need a more dynamic plan that allows them to follow hyperlinks and use forms with a preset goal. Therefore, in this article, we show how compositions of hypermedia APIs can be created by generic Semantic Web reasoners. This is achieved through the generation of a proof based on semantic descriptions of the APIs' functionality. To pragmatically verify the applicability of compositions, we introduce the notion of pre-execution and post-execution proofs. The runtime interaction between a client and a server is guided by proofs but driven by hypermedia, allowing the client to react to the application's actual state indicated by the server's response. We describe how to generate compositions from descriptions, discuss a computer-assisted process to generate descriptions, and verify reasoner performance on various composition tasks using a benchmark suite. The experimental results lead to the conclusion that proof-based consumption of hypermedia APIs is a feasible strategy at Web scale.

cs.LO↗

arXiv2014

Real-time animation of human characters with fuzzy controllers

The production of animation is a resource intensive process in game companies. Therefore, techniques to synthesize animations have been developed. However, these procedural techniques offer limited adaptability by animation artists. In order to solve this, a fuzzy neural network model of the animation is proposed, where the parameters can be tuned either by machine learning techniques that use motion capture data as training data or by the animation artist himself. This paper illustrates how this real time procedural animation system can be developed, taking the human gait on flat terrain and inclined surfaces as example. Currently, the parametric model is capable of synthesizing animations for various limb sizes and step sizes.

cs.GR↗