Refine
Document Type
- Conference Proceeding (213) (remove)
Conference Type
- Konferenzartikel (183)
- Konferenz-Abstract (21)
- Sonstiges (5)
- Konferenz-Poster (2)
- Konferenzband (2)
Keywords
- RoboCup (12)
- Machine Learning (9)
- Deep Leaning (7)
- Heart rhythm model (5)
- Modeling and simulation (5)
- Robustness (4)
- machine learning (4)
- Generative Adversarial Network (3)
- Radar (3)
- cryptography (3)
- neural networks (3)
- printed electronics (3)
- Computer Vision (2)
- Cryoballoon catheter ablation (2)
- Current measurement (2)
- Energy Flexibility (2)
- Geophysik (2)
- Kalman Filter (2)
- Mikroelektronik (2)
- NB-IoT (2)
- Neural networks (2)
- Predictive Maintenance (2)
- Security (2)
- Stability (2)
- Supraventricular tachycardia (2)
- Switches (2)
- UWB radars (2)
- accountability (2)
- artificial intelligence (2)
- atrial fibrillation (2)
- autoattack (2)
- cardiac ablation (2)
- catheter ablation (2)
- certificate management (2)
- convolutional neural networks (2)
- efficient training (2)
- explainability (2)
- fairness (2)
- heart rhythm model (2)
- image classification (2)
- imaging algorithms (2)
- interactive visualization (2)
- medical imaging (2)
- microwave (2)
- programming (2)
- responsibility (2)
- scattering measurements (2)
- semantics (2)
- trust (2)
- understandability (2)
- 3D print (1)
- 3D-Modelling (1)
- 5G (1)
- 5G mobile communication (1)
- 5G private networks (1)
- AC machines (1)
- AIN Cantilever (1)
- AV nodal reentry tachycardia (1)
- AV reentrant tachycardia (1)
- Additive Manufacturing (1)
- Adversarial Attacks (1)
- Adversarial Robustness (1)
- Agent based sensor (1)
- Air Pollution (1)
- Aliasing (1)
- Aluminum (1)
- Amplitude and Phase Errors (1)
- Angle of Arrival (1)
- Anti-Windup (1)
- Approximation (1)
- Artificial Intelligence (1)
- Atrial fibrillation (1)
- Authentication (1)
- Authorization (1)
- Automated idea generation (1)
- Automation (1)
- Automotive Radar (1)
- Battery storage (1)
- Bearings (1)
- Bioimpedance measurement (1)
- Biomimetics (1)
- Blockchain (1)
- Blockchains (1)
- Boundary conditions (1)
- Bowtie antenna (1)
- CNN (1)
- CNNs (1)
- Calibration (1)
- Cardiac Resynchronization Therapy (1)
- Cardiac resynchronization therapy (1)
- Challenges in Action Recognition (1)
- Chemical engineering (1)
- Cleaning (1)
- Clustering (1)
- Collaboration of Academia and Industry (1)
- Compensation (1)
- Control Algorithms (1)
- Current Control (1)
- DC-AC converters (1)
- Data Mining (1)
- Data breech (1)
- Debinding (1)
- Deep Learning (1)
- Deep Neural Network (1)
- Demand side flexibility (1)
- Device characterization (1)
- Diagnostics (1)
- Digital Beamforming (1)
- Digital Flex Twin Optimization (1)
- Digital Twin (1)
- Digitale Turschilder (1)
- Digitalization (1)
- Drehstrom (1)
- E-Ink (1)
- E-Paper (1)
- EAP-TLS (1)
- EKF-SLAM (1)
- ETAP Simulations (1)
- Eco-inventive principles (1)
- Economics (1)
- Edge AI (1)
- Education (1)
- Eigenvalues (1)
- Electromagnetic and thermal simulation (1)
- Embedded AI (1)
- Embedded Systems (1)
- Energiemanagement (1)
- Energy Flexibility for Companies (1)
- Energy Harvesting (1)
- Energy Management (1)
- Energy Marketing of Industrial Flexibilities (1)
- Energy management (1)
- Energy systems modeling (1)
- Environmental monitoring (1)
- Error (1)
- Esophageal catheter (1)
- Estimation (1)
- FUSION (1)
- Failure analysis (1)
- Fault Classification (1)
- Featherweight Go (1)
- Federated Learning (1)
- Field Programmable Gate Array (FPGA) (1)
- Fused Filament Fabrication (1)
- GPU Computing (1)
- Gleichstrom (1)
- Handschrift (1)
- Handswritten Character Recognition (1)
- Harmonic analysis (1)
- Heart Rhythm Simulation (1)
- Hemodynamic monitoring (1)
- Herzrhythmusmodell (1)
- High frequency ablation (1)
- His-Bundle Pacing (1)
- Hochspannung (1)
- Hybrid system (1)
- IEC/IEEE 60802 security (1)
- IEEE802.15.4 (1)
- IIoT (1)
- InceptionTime (1)
- Industrial Blockchain (1)
- Inertial (1)
- Ink (1)
- Internet of Things (1)
- Inventive principles (1)
- Inventive problem solving (1)
- Inverters (1)
- IoT Security (1)
- IoT security (1)
- Kryoballonkatheter (1)
- Künstliche Intelligenz (1)
- LPWAN (1)
- Large Grid-Connected PV Systems (1)
- Left Atrial Appendage Closure (1)
- Limiting (1)
- Load Flow Analysis (1)
- Logic gates (1)
- Long Term Evolution (1)
- Low-Cost Füllstandssensor (1)
- Low-Power (1)
- MEMS (1)
- MIMO (1)
- MLOps (1)
- Machine learning (1)
- Magnetic sensors (1)
- Manufacturing automation (1)
- Manufacturing industries (1)
- Mean Square Error (1)
- Measurement (1)
- Medizintechnik (1)
- Microgrid(s) (1)
- Mode Collapse (1)
- Model Calibration (1)
- Model Predictive Control (1)
- Model Search (1)
- Modellierung und Simulation (1)
- Monitoring (1)
- Monocular Depth Estimation (1)
- Monte-Carlo (1)
- Monte-Carlo Simulation (1)
- Monte-Carlo method (1)
- Multiside heart stimulation (1)
- NETCONF security (1)
- NFC (1)
- Nature-inspired principles (1)
- Network Test (1)
- Netzintegration (1)
- Nyquist-Shannon (1)
- OT security (1)
- Octave Convolution (1)
- Optimization (1)
- Optimization and control (1)
- Optimization with Digital Twins (1)
- Optische Zeichenerkennung (1)
- PI control (1)
- PKI (1)
- PROFINET IO (1)
- PROFINET Security (1)
- PV System (1)
- Parallelization (1)
- Parameter Estimation (1)
- Parameter estimation (1)
- Pattern Recognition (1)
- Peak shaving (1)
- Peer to peer network (1)
- Performance (1)
- Performance evaluation (1)
- Permanent magnet machines (1)
- Physical Unclonable Functions (1)
- Physical unclonable function (1)
- Physiological Pacing (1)
- Power Loss (1)
- Predictions (1)
- Predictive Models (1)
- Process design (1)
- Pulmonalvenenisolation (1)
- Pulmonary vein isolation (1)
- Pulse width modulation (1)
- RFID (1)
- RUL (1)
- Radio frequency (1)
- Random call model (1)
- Real-Time Communication (1)
- Regularization (1)
- Renewable Energy Markets (1)
- Representation Learning (1)
- ResNet (1)
- Road-Quality Prediction (1)
- Roboter (1)
- Rotors (1)
- Sampling (1)
- Second-order Optimization (1)
- Self-Calibration (1)
- Semiconductor Device (1)
- Semiconductor device measurement (1)
- Sensor phenomena and characterization (1)
- Sintering (1)
- Smart Energy Metering (1)
- Smart Gardening (1)
- Smart Grids (1)
- Smart Home (1)
- Smart-UPS (1)
- SmartMAC (1)
- Software (1)
- Software algorithms (1)
- Software for measurements (1)
- Solar Radiation (1)
- Spinal cord stimulation (1)
- Stromregelung (1)
- Stromzustandsregler (1)
- Subspace Clustering (1)
- Substrates (1)
- Sustainable technology (1)
- Synchronmaschine (1)
- Systematic innovation (1)
- TLS (1)
- TRIZ methodology (1)
- TSN security (1)
- Testbed (1)
- Time Sensitive Networking (1)
- Time Synchronization (1)
- Time series data (1)
- Time-series Classification (1)
- TinyML (1)
- Torque (1)
- Total Harmonic Distortion (1)
- Traceability (1)
- Training (1)
- Transesophageal left atrial pacing (1)
- Trust management (1)
- Ultra-Low Energy (1)
- Unsupervised Conditional Training (1)
- Unsupervised Learning (1)
- Variational Autoencoders (1)
- Virtual Reality (1)
- Vorhofflimmern (1)
- Wireless IoT (1)
- Wireless Sensor Nodes (1)
- Wärmepumpen (1)
- Yolov5 (1)
- accelerometer (1)
- adversarial (1)
- adversarial attacks (1)
- adversarial detection (1)
- atrial flutter (1)
- attribute manipulation (1)
- autoML (1)
- autonomous systems (1)
- bearing (1)
- bench-marking (1)
- benchmarking (1)
- biocompatibility test (1)
- biodegradable (1)
- biomaterials (1)
- biomechanical stimulation (1)
- building management systems (1)
- camera-based navigation (1)
- cellular radio (1)
- cifar (1)
- computer network management (1)
- correlation (1)
- credentials (1)
- crossbar (1)
- curb (1)
- curriculum learning (1)
- cybersecurity (1)
- deep learning (1)
- deep reinforcement learning (1)
- defense (1)
- degradation stages (1)
- detection (1)
- dictionary passing (1)
- echocardiography (1)
- electrolyte-gated transistors (1)
- embedded systems (1)
- energy harvesting (1)
- face editing (1)
- face recognition (1)
- fail-operational (1)
- fingerprinting (1)
- fourier (1)
- gan (1)
- generative adversarial networks (1)
- ground penetrating radar (1)
- hair (1)
- heart rhythm simulation (1)
- heat pump (1)
- height estimation (1)
- home automation (1)
- hybrid systems (1)
- identification (1)
- image color analysis (1)
- imagenet (1)
- impedance cardiography (1)
- industrial Ethernet (1)
- industrial IoT (1)
- industrial communication (1)
- irrigation (1)
- lid (1)
- machine-to-machine communication (1)
- mahalanobis (1)
- metal oxide transistor (1)
- millimeter-wave (1)
- model-predictive control (1)
- molybdenum (1)
- multipath (1)
- mutual authentication (1)
- natürliche Leistung (1)
- neural architecture search (1)
- noise measurement (1)
- nose (1)
- parking (1)
- pattern recognition (1)
- perception (1)
- performance (1)
- physical unclonable function (1)
- physically unclonable function (PUF) (1)
- predictive maintenance (1)
- primary authentication (1)
- printed Antennas (1)
- programming languages (1)
- pruning (1)
- pulmonary vein isolation (1)
- radio networks (1)
- rekeying (1)
- resource efficiency (1)
- secure communication (1)
- security (1)
- skin cancer (1)
- skin cancer detection (1)
- software defined radio (1)
- solar module (1)
- sparse backpropagation (1)
- spectral defense (1)
- spectraldefense (1)
- statistical methods, ROS (1)
- stochastic computing (1)
- style (1)
- style transfer (1)
- system authenticity (1)
- telecommunication equipment testing (1)
- temperature sensor (1)
- thinned ASIC in foil (1)
- transmit beamforming (1)
- ventricular tachycardia (1)
- wide area networks (1)
- wireless sensor networks (1)
- übertragbare Wirkleistung (1)
Institute
- Fakultät Elektrotechnik, Medizintechnik und Informatik (EMI) (ab 04/2019) (213) (remove)
Open Access
- Closed Access (82)
- Open Access (75)
- Closed (55)
- Bronze (27)
- Diamond (9)
- Grün (3)
- Gold (1)
In many application areas, Deep Reinforcement Learning (DRL) has led to breakthroughs. In Curriculum Learning, the Machine Learning algorithm is not randomly presented with examples, but in a meaningful order of increasing difficulty. This has been used in many application areas to further improve the results of learning systems or to reduce their learning time. Such approaches range from learning plans created manually by domain experts to those created automatically. The automated creation of learning plans is one of the biggest challenges.In this work, we investigate an approach in which a trainer learns in parallel and analogously to the student to automatically create a learning plan for the student during this Double Deep Reinforcement Learning (DDRL). Three Reward functions, Friendly, Adversarial, and Dynamic based on the learner’s reward are compared. The domain for evaluation is kicking with variable distance, direction and relative ball position in the SimSpark simulated soccer environment.As a result, Statistic Curriculum Learning (SCL) performs better than a random curriculum with respect to training time and result quality. DDRL reaches a comparable quality as the baseline and outperforms it significantly in shorter trainings in the distance-direction subdomain reducing the number of required training cycles by almost 50%.
Team description papers of magmaOffenburg are incremental in the sense that each year we address a different topic of our team and the tools around our team. In this year’s team description paper we focus on the architecture of the software. It is a main factor for being able to keep the code maintainable even after 15 years of development. We also describe how we make sure that the code follows this architecture.
Ensuring that software applications present their users the most recent version of data is not trivial. Self-adjusting computations are a technique for automatically and efficiently recomputing output data whenever some input changes.
This article describes the software architecture of a large, commercial software system built around a framework for coarse-grained self-adjusting computations in Haskell. It discusses advantages and disadvantages based on longtime experience. The article also presents a demo of the system and explains the API of the framework.
Detecting Images Generated by Deep Diffusion Models using their Local Intrinsic Dimensionality
(2023)
Diffusion models recently have been successfully applied for the visual synthesis of strikingly realistic appearing images. This raises strong concerns about their potential for malicious purposes. In this paper, we propose using the lightweight multi Local Intrinsic Dimensionality (multiLID), which has been originally developed in context of the detection of adversarial examples, for the automatic detection of synthetic images and the identification of the according generator networks. In contrast to many existing detection approaches, which often only work for GAN-generated images, the proposed method provides close to perfect detection results in many realistic use cases. Extensive experiments on known and newly created datasets demonstrate that the proposed multiLID approach exhibits superiority in diffusion detection and model identification.Since the empirical evaluations of recent publications on the detection of generated images are often mainly focused on the "LSUN-Bedroom" dataset, we further establish a comprehensive benchmark for the detection of diffusion-generated images, including samples from several diffusion models with different image sizes.The code for our experiments is provided at https://github.com/deepfake-study/deepfake-multiLID.
Erlang is a functional programming language with dynamic typing. The language offers great flexibility for destructing values through pattern matching and dynamic type tests. Erlang also comes with a type language supporting parametric polymorphism, equi-recursive types, as well as union and a limited form of intersection types. However, type signatures only serve as documentation; there is no check that a function body conforms to its signature.
Set-theoretic types and semantic subtyping fit Erlang’s feature set very well. They allow expressing nearly all constructs of its type language and provide means for statically checking type signatures. This article brings set-theoretic types to Erlang and demonstrates how existing Erlang code can be statically type checked without or with only minor modifications to the code. Further, the article formalizes the main ingredients of the type system in a small core calculus, reports on an implementation of the system, and compares it with other static type checkers for Erlang.
In recent years, predictive maintenance tasks, especially for bearings, have become increasingly important. Solutions for these use cases concentrate on the classification of faults and the estimation of the Remaining Useful Life (RUL). As of today, these solutions suffer from a lack of training samples. In addition, these solutions often require high-frequency accelerometers, incurring significant costs. To overcome these challenges, this research proposes a combined classification and RUL estimation solution based on a Convolutional Neural Network (CNN) and a Long Short-Term Memory (LSTM) network. This solution relies on a hybrid feature extraction approach, making it especially appropriate for low-cost accelerometers with low sampling frequencies. In addition, it uses transfer learning to be suitable for applications with only a few training samples.
TSN, or Time Sensitive Networking, is becoming an essential technology for integrated networks, enabling deterministic and best effort traffic to coexist on the same infrastructure. In order to properly configure, run and secure such TSN, monitoring functionality is a must. The TSN standard already has some preparations to provide such functionality and there are different methods to choose from. We implemented different methods to measure the time synchronisation accuracy between devices as a C library and compared the measurement results. Furthermore, the library has been integrated into the ControlTSN engineering framework.
Neural networks have a number of shortcomings. Amongst the severest ones is the sensitivity to distribution shifts which allows models to be easily fooled into wrong predictions by small perturbations to inputs that are often imperceivable to humans and do not have to carry semantic meaning. Adversarial training poses a partial solution to address this issue by training models on worst-case perturbations. Yet, recent work has also pointed out that the reasoning in neural networks is different from humans. Humans identify objects by shape, while neural nets mainly employ texture cues. Exemplarily, a model trained on photographs will likely fail to generalize to datasets containing sketches. Interestingly, it was also shown that adversarial training seems to favorably increase the shift toward shape bias. In this work, we revisit this observation and provide an extensive analysis of this effect on various architectures, the common L_2-and L_-training, and Transformer-based models. Further, we provide a possible explanation for this phenomenon from a frequency perspective.
Seismic data processing relies on multiples attenuation to improve inversion and interpretation. Radon-based algorithms are often used for multiples and primaries discrimination. Deep learning, based on convolutional neural networks (CNNs), has shown encouraging applications for demultiple that could mitigate Radon-based challenges. In this work, we investigate new strategies to train a CNN for multiples removal based on different loss functions. We propose combined primaries and multiples labels in the loss for training a CNN to predict primaries, multiples, or both simultaneously. Moreover, we investigate two distinctive training methods for all the strategies: UNet based on minimum absolute error (L1) training, and adversarial training (GAN-UNet). We test the trained models with the different strategies and methods on 400 synthetic data. We found that training to predict multiples, including the primaries …
Sweaty has already participated several times in RoboCup soccer competitions (Adult Size). Now the work is focused coordinating the play of two robots. Moreover, we are working on stabilizing the gait by adding additional sensor information. An ongoing work is the optimization of the control strategy by balancing between impedance and position control. By minimizing the jerk, gait and overall gameplay should improve significantly.