Refine
Year of publication
- 2022 (97) (remove)
Document Type
- Conference Proceeding (97) (remove)
Conference Type
- Konferenzartikel (82)
- Konferenz-Abstract (11)
- Konferenz-Poster (2)
- Sonstiges (2)
Language
- English (97) (remove)
Is part of the Bibliography
- yes (97)
Keywords
- injury (10)
- Machine Learning (5)
- biomechanics (5)
- running (5)
- ACL (4)
- Robustness (4)
- Radar (3)
- RoboCup (3)
- sport (3)
- 3D printing (2)
Institute
- Fakultät Elektrotechnik, Medizintechnik und Informatik (EMI) (ab 04/2019) (40)
- Fakultät Maschinenbau und Verfahrenstechnik (M+V) (26)
- Fakultät Wirtschaft (W) (20)
- Fakultät Medien (M) (ab 22.04.2021) (13)
- INES - Institut für nachhaltige Energiesysteme (12)
- ivESK - Institut für verlässliche Embedded Systems und Kommunikationselektronik (11)
- IMLA - Institute for Machine Learning and Analytics (10)
- ACI - Affective and Cognitive Institute (3)
- IUAS - Institute for Unmanned Aerial Systems (3)
- POIM - Peter Osypka Institute of Medical Engineering (2)
Open Access
- Open Access (47)
- Closed (41)
- Bronze (26)
- Diamond (11)
- Closed Access (9)
- Grün (6)
- Hybrid (2)
- Gold (1)
The isolation measures adopted during the COVID-19 pandemic brought light to discussions related to the importance of meaningful social relationships as a basic need to human well-being. But even before the pandemic outbreak in the years 2020 and 2021, organizations and scholars were already drawing attention to the growing numbers related to lonely people in the world (World Economic Forum, 2019). Loneliness is an emotional distress caused by the lack of meaningful social connections, which affects people worldwide across all age groups, mainly young adults (Rook, 1984). The use of digital technologies has gained prominence as a means of alleviating the distress. As an example, studies have shown the benefits of using digital games both to stimulate social interactions (Steinfield, Ellison & Lampe, 2008) and to enhance the effects of digital interventions for mental health treatments, through gamification (Fleming et al., 2017). It is with these aspects in mind that the gamified app Noneliness was designed with the intention of reducing loneliness rates among young students at a German university. In addition to sharing the related works that supported the application development, this chapter also presents the aspects considered for the resource's design, its main functionalities, and the preliminary results related to the reduction of loneliness in the target audience.
We aim to debate and eventually be able to carefully judge how realistic the following statement of a young computer scientist is: “I would like to become an ethical correctly acting offensive cybersecurity expert”. The objective of this article is not to judge what is good and what is wrong behavior nor to present an overall solution to ethical dilemmas. Instead, the goal is to become aware of the various personal moral dilemmas a security expert may face during his work life. For this, a total of 14 cybersecurity students from HS Offenburg were asked to evaluate several case studies according to different ethical frameworks. The results and particularities are discussed, considering different ethical frameworks. We emphasize, that different ethical frameworks can lead to different preferred actions and that the moral understanding of the frameworks may differ even from student to student.
Sweaty has already participated several times in RoboCup soccer competitions (Adult Size). Now the work is focused on stabilizing the gait. Moreover, we would like to overcome the constraints of a ZMP-algorithm that has a horizontal footplate as precondition for the simplification of the equations. In addition we would like to switch between impedance and position control with a fuzzy-like algorithm that might help to minimize jerks when Sweaty’s feet touch the ground.
Generative machine learning models for creative purposes play an increasingly prominent role in the field of dance and technology. A particularly popular approach is the use of such models for generating synthetic motions. Such motions can either serve as source of ideation for choreographers or control an artificial dancer that acts as improvisation partner for human dancers. Several examples employ autoencoder-based deep-learning architectures that have been trained on motion capture recordings of human dancers. Synthetic motions are then generated by navigating the autoencoder's latent space. This paper proposes an alternative approach of using an autoencoder for creating synthetic motions. This approach controls the generation of synthetic motions on the level of the motion itself rather than its encoding. Two different methods are presented that follow this principle. Both methods are based on the interactive control of a single joint of an artificial dancer while the other joints remain under the control of the autoencoder. The first method combines the control of the orientation of a joint with iterative autoencoding. The second method combines the control of the target position of a joint with forward kinematics and the application of latent difference vectors. As illustrative example of an artistic application, this latter method is used for an artificial dancer that plays a digital instrument. The paper presents the implementation of these two methods and provides some preliminary results.
Subspace clustering aims to find all clusters in all subspaces of a high-dimensional data space. We present a massively data-parallel approach that can be run on graphics processing units. It extends a previous density-based method that scales well with the number of dimensions. Its main computational bottleneck consists of (sequentially) generating a large number of minimal cluster candidates in each dimension and using hash collisions in order to find matches of such candidates across multiple dimensions. Our approach parallelizes this process by removing previous interdependencies between consecutive steps in the sequential generation process and by applying a very efficient parallel hashing scheme optimized for GPUs. This massive parallelization gives up to 70x speedup for
the bottleneck computation when it is replaced by our approach and run on current GPU hardware. We note that depending on data size and choice of parameters, the parallelized part of the algorithm can take different percentages of the overall runtime of the clustering process, and thus, the overall clustering speedup may vary significantly between different cases. However, even
in our ”worst-case” test, a small dataset where the computation makes up only a small fraction of the overall clustering time, our parallel approach still yields a speedup of more than 3x for the complete run of the clustering process. Our method could also be combined with parallelization of other parts of the clustering algorithm, with an even higher potential gain in processing speed.
Brand-related-user-generated-content allows companies to achieve several important objectives, such as increasing sales and creating higher user engagement. In this paper a research framework is developed that provides an overview of the necessary processes to successfully use brand-related-user-generated-content. The framework also helps managers to understand the main motives of users when posting brand-related-user-generated-content. Expert interviews were carried out to validate the research framework. The results from the interviews support the proposed framework. Brand-related-user-generated-content can increase purchase intention and the community engagement. From a user’s perspective the opportunity to interact with a brand and be featured on official brand channels could be seen as the main motivation for creating brand-related-user-generated-content.
Significant improvements in module performance are possible via implementation of multi-wire electrodes. This is economically sound as long as the mechanical yield of the production is maintained. While flat ribbons have a relatively large contact area to exert forces onto the solar cell, wires with round cross section reduce this contact area considerably – in theory to an infinitively thin line. Therefore, the local stresses induced by the electrodes might increase to a point that mechanical production yields suffer unacceptably.
In this paper, we assess this issue by an analytical mechanical model as well as experiments with an encapsulant-free N.I.C.E. test setup. From these, we can derive estimations for the relationship between lay-up accuracy and expected breakage losses. This paves the way for cost-optimized choices of handling equipment in industrial N.I.C.E.-wire production lines.
Micronization of biochar (BC) may ease its application in agriculture. For example, fine biochar powders can be applied as suspensions via drip-irrigation systems or can be used to produce grnulated fertilizers. However, micronization may effect important physical biochar properties like the water holding capacity (WHC) or the porosity.
The aim of this study is to identify indicators at country level that could prove useful in improving the effectiveness of fraud detection in European Structural and Investment Funds. The chapter analyses EU funds, belonging to the period 2014–2020, from and the study suggests the convenience of tracking funds, especially in countries with higher GDP and higher transparency levels, and the lesser relevance of the number of irregularities for countries with higher GDP and those receiving larger funds. Fraud and fraud detection rates in individual funds vary significantly across states. Federal states, such as the Federal Republic of Germany, are comparatively successful in detecting fraud in EU funds.
Currently, many theoretical as well as practically relevant questions towards the transferability and robustness of Convolutional Neural Networks (CNNs) remain unsolved. While ongoing research efforts are engaging these problems from various angles, in most computer vision related cases these approaches can be generalized to investigations of the effects of distribution shifts in image data. In this context, we propose to study the shifts in the learned weights of trained CNN models. Here we focus on the properties of the distributions of dominantly used 3×3 convolution filter kernels. We collected and publicly provide a dataset with over 1.4 billion filters from hundreds of trained CNNs, using a wide range of datasets, architectures, and vision tasks. In a first use case of the proposed dataset, we can show highly relevant properties of many publicly available pre-trained models for practical applications: I) We analyze distribution shifts (or the lack thereof) between trained filters along different axes of meta-parameters, like visual category of the dataset, task, architecture, or layer depth. Based on these results, we conclude that model pre-training can succeed on arbitrary datasets if they meet size and variance conditions. II) We show that many pre-trained models contain degenerated filters which make them less robust and less suitable for fine-tuning on target applications. Data & Project website: https://github.com/paulgavrikov/cnn-filter-db.
Over the last years, Convolutional Neural Networks (CNNs) have been the dominating neural architecture in a wide range of computer vision tasks. From an image and signal processing point of view, this success might be a bit surprising as the inherent spatial pyramid design of most CNNs is apparently violating basic signal processing laws, i.e. Sampling Theorem in their down-sampling operations. However, since poor sampling appeared not to affect model accuracy, this issue has been broadly neglected until model robustness started to receive more attention. Recent work in the context of adversarial attacks and distribution shifts, showed after all, that there is a strong correlation between the vulnerability of CNNs and aliasing artifacts induced by poor down-sampling operations. This paper builds on these findings and introduces an aliasing free down-sampling operation which can easily be plugged into any CNN architecture: FrequencyLowCut pooling. Our experiments show, that in combination with simple and Fast Gradient Sign Method (FGSM) adversarial training, our hyper-parameter free operator substantially improves model robustness and avoids catastrophic overfitting. Our code is available at https://github.com/GeJulia/flc_pooling
Deep learning models are intrinsically sensitive to distribution shifts in the input data. In particular, small, barely perceivable perturbations to the input data can force models to make wrong predictions with high confidence. An common defense mechanism is regularization through adversarial training which injects worst-case perturbations back into training to strengthen the decision boundaries, and to reduce overfitting. In this context, we perform an investigation of 3 × 3 convolution filters that form in adversarially- trained models. Filters are extracted from 71 public models of the ℓ ∞ -RobustBench CIFAR-10/100 and ImageNet1k leaderboard and compared to filters extracted from models built on the same architectures but trained without robust regularization. We observe that adversarially-robust models appear to form more diverse, less sparse, and more orthogonal convolution filters than their normal counterparts. The largest differences between robust and normal models are found in the deepest layers, and the very first convolution layer, which consistently and predominantly forms filters that can partially eliminate perturbations, irrespective of the architecture.
Despite the success of convolutional neural networks (CNNs) in many academic benchmarks for computer vision tasks, their application in the real-world is still facing fundamental challenges. One of these open problems is the inherent lack of robustness, unveiled by the striking effectiveness of adversarial attacks. Current attack methods are able to manipulate the network's prediction by adding specific but small amounts of noise to the input. In turn, adversarial training (AT) aims to achieve robustness against such attacks and ideally a better model generalization ability by including adversarial samples in the trainingset. However, an in-depth analysis of the resulting robust models beyond adversarial robustness is still pending. In this paper, we empirically analyze a variety of adversarially trained models that achieve high robust accuracies when facing state-of-the-art attacks and we show that AT has an interesting side-effect: it leads to models that are significantly less overconfident with their decisions, even on clean data than non-robust models. Further, our analysis of robust models shows that not only AT but also the model's building blocks (like activation functions and pooling) have a strong influence on the models' prediction confidences. Data & Project website: https://github.com/GeJulia/robustness_confidences_evaluation
Estimating the Robustness of Classification Models by the Structure of the Learned Feature-Space
(2022)
Over the last decade, the development of deep image classification networks has mostly been driven by the search for the best performance in terms of classification accuracy on standardized benchmarks like ImageNet. More recently, this focus has been expanded by the notion of model robustness, \ie the generalization abilities of models towards previously unseen changes in the data distribution. While new benchmarks, like ImageNet-C, have been introduced to measure robustness properties, we argue that fixed testsets are only able to capture a small portion of possible data variations and are thus limited and prone to generate new overfitted solutions. To overcome these drawbacks, we suggest to estimate the robustness of a model directly from the structure of its learned feature-space. We introduce robustness indicators which are obtained via unsupervised clustering of latent representations from a trained classifier and show very high correlations to the model performance on corrupted test data.
Many commonly well-performing convolutional neural network models have shown to be susceptible to input data perturbations, indicating a low model robustness. Adversarial attacks are thereby specifically optimized to reveal model weaknesses, by generating small, barely perceivable image perturbations that flip the model prediction. Robustness against attacks can be gained for example by using adversarial examples during training, which effectively reduces the measurable model attackability. In contrast, research on analyzing the source of a model’s vulnerability is scarce. In this paper, we analyze adversarially trained, robust models in the context of a specifically suspicious network operation, the downsampling layer, and provide evidence that robust models have learned to downsample more accurately and suffer significantly less from aliasing than baseline models.
The authors set the focus in this paper on the description of polarization with the help of the Jones calculus and the application of polarization in photography. Furthermore, the effect of the circular polarization filter is described by using the Jones calculus. Also, an enhancement of artistic and creative possibilities in photography through quantization or parametrization by the Jones matrices is presented.
Teaching and learning concepts that are adapted to the constantly evolving requirements due to rapid technological progress are essential for teaching in media photonics technology. After the development of a concept for research-oriented education in optics and photonics, the next step will be a conceptual restructuring and redesign of the entire curriculum for education in media photonics technology. By including typical research activities as essential components of the learning process, a broad platform for practical projects and applied research can be created, offering a variety of new development opportunities.