Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
1,247
result(s) for
"Peters, Steven"
Sort by:
Before the fall : German and Austrian art of the 1930s
This exhibition, comprised of nearly 150 paintings and works on paper, will trace the many routes traveled by German and Austrian artists and will demonstrate the artistic developments that foreshadowed, reflected, and accompanied the beginning of World War II. Central topics of the exhibition will be the reaction of the artists towards their historical circumstances, the development of style with regard to the appropriation of various artistic idioms, the personal fate of artists, and major political events that shaped the era. The show assembles key works by leading artists such as Max Beckmann, Otto Dix, Max Ernst, Oskar Kokoschka, and Alfred Kubin, and artists less familiar to audiences in the United States including Friedl Dicker-Brandeis, Albert Paris Gèutersloh, Karl Hubbuch, Richard Oelze, Franz Sedlacek, Josef Scharl, and Rudolf Wacker, who will each be represented by small groups of significant works. Among the important loans to the exhibition will be Max Beckmann's \"Bird Hell\" from 1937-38, Oskar Kokoschka's \"Portrait of Thomas G. Masaryk\" from 1935-36, and Richard Oelze's \"Uncanny Expectation\" from 1935-36. The exhibition will also feature photographic portraits by Helmar Lerski and August Sander. 00Exhibition: Neue Galerie, New York, USA (08.03. - 28.05.2018).
Book
Towards CRISP-ML(Q): A Machine Learning Process Model with Quality Assurance Methodology
Machine learning is an established and frequently used technique in industry and academia, but a standard process model to improve success and efficiency of machine learning applications is still missing. Project organizations and machine learning practitioners face manifold challenges and risks when developing machine learning applications and have a need for guidance to meet business expectations. This paper therefore proposes a process model for the development of machine learning applications, covering six phases from defining the scope to maintaining the deployed machine learning application. Business and data understanding are executed simultaneously in the first phase, as both have considerable impact on the feasibility of the project. The next phases are comprised of data preparation, modeling, evaluation, and deployment. Special focus is applied to the last phase, as a model running in changing real-time environments requires close monitoring and maintenance to reduce the risk of performance degradation over time. With each task of the process, this work proposes quality assurance methodology that is suitable to address challenges in machine learning development that are identified in the form of risks. The methodology is drawn from practical experience and scientific literature, and has proven to be general and stable. The process model expands on CRISP-DM, a data mining process model that enjoys strong industry support, but fails to address machine learning specific tasks. The presented work proposes an industry- and application-neutral process model tailored for machine learning applications with a focus on technical tasks for quality assurance.
Journal Article
Physics-informed attention on temporal fusion transformer for multivariate truck range forecasting
Accurate real-time forecasting of remaining vehicle range remains a challenge, particularly under varying payloads, road gradients, and environmental conditions. Conventional Temporal Fusion Transformers (TFTs) utilize attention mechanisms to dynamically weight historical inputs but may fail to capture explicit physical relationships that are critical for accurate predictions in heavy-duty electric trucks. This paper introduces a novel approach to integrating physical vehicle dynamics directly into the attention mechanism of TFTs. Our method, Physics-Informed Attention for TFT (PIA-TFT), modifies attention calculation by injecting physics-based relevance scores derived from vehicle speed, payload, road gradients, and other physical parameters, improving interpretability and model accuracy under operational conditions. Empirical evaluations conducted with real-world data from electric trucks demonstrate that the PIA-TFT reduces prediction errors compared to standard TFTs by up to 18%. Our approach is a step towards more physically consistent and explainable deep learning architectures for automotive forecasting tasks.
Journal Article
Energy-efficient traffic sign recognition using directly trained spiking neural networks and population decoding
Recognizing traffic signs is a fundamental perception task for automated driving systems and requires high accuracy under strict latency and energy constraints. Convolutional neural networks (CNNs) achieve strong performance but can be computationally demanding for embedded platforms. Spiking convolutional neural networks (SCNNs) offer an event-driven alternative that can reduce computation through sparse activity, yet their accuracy often degrades under very low-latency settings with few time steps. To improve spike-based inference under strict runtime constraints, we integrate a neural population decoding layer at the output stage and evaluate directly trained SCNNs with and without population decoding against a CNN baseline on the German Traffic Sign Recognition Benchmark (GTSRB). The best SCNN without population decoding achieved 98.85% test accuracy at 30 time steps, exceeding the CNN baseline of 98.38%. Population decoding improved performance in the low-latency regime, reaching 98.31% accuracy at a single time step, corresponding to an improvement of 0.56% over the SCNN without population decoding at the same temporal setting. Using an operation-based energy estimation, the SCNNs achieved over 14 times higher energy efficiency than the CNN at one time step. Overall, the results demonstrate that directly trained SCNNs can surpass a comparable CNN while enabling flexible trade-offs between accuracy, inference time, and energy efficiency. In particular, population decoding proves beneficial when operating under strict latency constraints.
Journal Article
Basilar artery stenting in hyperacute stroke: A systematic review of published cases
Basilar artery stenting is a rescue therapy in the management of hyperacute stroke. Published data on efficacy and safety are limited.
A systematic review of published studies was performed in accordance with PRISMA guidelines. Inclusion criteria were adult patients with ischemic stroke with permanent basilar artery stent placement within 48 h of onset. Data were extracted by two independent reviewers. Additional cases from our institution were identified via a local stroke registry.
Of 212 screened articles, patient-level data was reported in 35 studies (87 individuals) and six additional patients were included from our registry. Patients (n = 93, 63 % male; median age 64) most often presented with mid-basilar occlusion (52 %) and 76 % received treatment within 12 hours of onset. Favorable angiographic results occurred in 67 %. The final modified Rankin Scale score (mRS) was 0–3 for 56 % of patients; mortality was 29 %. Those with complete flow post-procedure were more likely to have a final mRS of 0–3 (p = 0.05).
In 93 cases of basilar stenting in hyperacute stroke, favourable angiographic and functional outcomes were reported in 67 % and 56 % of patients, respectively. International multicenter registries are required to establish benefit and identify patient and technical factors that predict favorable outcomes.
•The use of unplanned basilar artery stenting during emergent stroke treatment occurs rarely but has not been well reported in the literature.•Published case series of patient level outcomes show good clinical outcome in over 50 % successful recanalization in 2/3 of cases.•Improved functional outcomes were recorded after 2016
Journal Article
Introducing a Development Method for Active Perception Sensor Simulations Using Continuous Verification and Validation
Simulation-based testing is playing an increasingly important role in the development and validation of automated driving functions, as real-world testing is often limited by cost, safety, and scalability. An essential part of this is the simulation of active perception sensors such as lidar and radar which enable accurate perception of the vehicle’s environment. In this context, the particular challenge lies in ensuring the credibility of these sensor simulations. This paper presents a novel method for the efficient and credible realization and validation of active perception sensor simulations in the context of the overall development process. Since the validity of these simulations is crucial for the safety augmentation of automated driving functions, the proposed method integrates a continuous verification and validation approach into the development process. Using this method, requirements like individual sensor effects are iteratively implemented into the simulation. Every iteration ends with the verification and validation of the resulting simulation. In addition, initial practical approaches are presented for validating measurement data required for the development process to avoid errors in data acquisition and for deriving quantified acceptance criteria as part of the validation process. All new approaches and methods are subsequently demonstrated on the example of a ray tracing-based lidar sensor simulation.
Journal Article
Cyclic and Sleep-Like Spontaneous Alternations of Brain State Under Urethane Anaesthesia
Although the induction of behavioural unconsciousness during sleep and general anaesthesia has been shown to involve overlapping brain mechanisms, sleep involves cyclic fluctuations between different brain states known as active (paradoxical or rapid eye movement: REM) and quiet (slow-wave or non-REM: nREM) stages whereas commonly used general anaesthetics induce a unitary slow-wave brain state.
Long-duration, multi-site forebrain field recordings were performed in urethane-anaesthetized rats. A spontaneous and rhythmic alternation of brain state between activated and deactivated electroencephalographic (EEG) patterns was observed. Individual states and their transitions resembled the REM/nREM cycle of natural sleep in their EEG components, evolution, and time frame ( approximately 11 minute period). Other physiological variables such as muscular tone, respiration rate, and cardiac frequency also covaried with forebrain state in a manner identical to sleep. The brain mechanisms of state alternations under urethane also closely overlapped those of natural sleep in their sensitivity to cholinergic pharmacological agents and dependence upon activity in the basal forebrain nuclei that are the major source of forebrain acetylcholine. Lastly, stimulation of brainstem regions thought to pace state alternations in sleep transiently disrupted state alternations under urethane.
Our results suggest that urethane promotes a condition of behavioural unconsciousness that closely mimics the full spectrum of natural sleep. The use of urethane anaesthesia as a model system will facilitate mechanistic studies into sleep-like brain states and their alternations. In addition, it could also be exploited as a tool for the discovery of new molecular targets that are designed to promote sleep without compromising state alternations.
Journal Article
Making Automotive Radar Sensor Validation Measurements Comparable
Virtual validation of radar sensor models is becoming increasingly important for the safety validation of Light Detection and Rangings (lidars). Therefore, methods for quantitative comparison of radar measurements in the context of model validation need to be developed. This paper presents a novel methodology for accessing and quantifying validation measurements of radar sensor models. This method uses Light Detection and Rangings (lidars) and the so-called Double Validation Metric (DVM) to effectively quantify deviations between distributions. By applying this metric, the study measures the reproducibility and repeatability of radar sensor measurements. Different interfaces and different levels of detail are investigated. By comparing the radar signals from real-world experiments where different objects are present, valuable insights are gained into the performance of the sensor. In particular, the research extends to assessing the impact of varying rain intensities on the measurement results, providing a comprehensive understanding of the sensor’s behavior under these conditions. This holistic approach significantly advances the evaluation of radar sensor capabilities and enables the quantification of the maximum required quality of radar simulation models.
Journal Article
Relay Architectures for 3GPP LTE-Advanced
The Third Generation Partnership Project's Long Term Evolution-Advanced is considering relaying for cost-effective throughput enhancement and coverage extension. While analog repeaters have been used to enhance coverage in commercial cellular networks, the use of more sophisticated fixed relays is relatively new. The main challenge faced by relay deployments in cellular systems is overcoming the extra interference added by the presence of relays. Most prior work on relaying does not consider interference, however. This paper analyzes the performance of several emerging half-duplex relay strategies in interference-limited cellular systems: one-way, two-way, and shared relays. The performance of each strategy as a function of location, sectoring, and frequency reuse are compared with localized base station coordination. One-way relaying is shown to provide modest gains over single-hop cellular networks in some regimes. Shared relaying is shown to approach the gains of local base station coordination at reduced complexity, while two-way relaying further reduces complexity but only works well when the relay is close to the handset. Frequency reuse of one, where each sector uses the same spectrum, is shown to have the highest network throughput. Simulations with realistic channel models provide performance comparisons that reveal the importance of interference mitigation in multihop cellular networks.
Journal Article