TY  - JOUR
A1  - Belaid, Mohamed Karim
A1  - Rabus, Maximilian
A1  - Krestel, Ralf
T1  - CrashNet
BT  - an encoder-decoder architecture to predict crash test outcomes
JF  - Data mining and knowledge discovery
N2  - Destructive car crash tests are an elaborate, time-consuming, and expensive necessity of the automotive development process. Today, finite element method (FEM) simulations are used to reduce costs by simulating car crashes computationally. We propose CrashNet, an encoder-decoder deep neural network architecture that reduces costs further and models specific outcomes of car crashes very accurately. We achieve this by formulating car crash events as time series prediction enriched with a set of scalar features. Traditional sequence-to-sequence models are usually composed of convolutional neural network (CNN) and CNN transpose layers. We propose to concatenate those with an MLP capable of learning how to inject the given scalars into the output time series. In addition, we replace the CNN transpose with 2D CNN transpose layers in order to force the model to process the hidden state of the set of scalars as one time series. The proposed CrashNet model can be trained efficiently and is able to process scalars and time series as input in order to infer the results of crash tests. CrashNet produces results faster and at a lower cost compared to destructive tests and FEM simulations. Moreover, it represents a novel approach in the car safety management domain.
KW  - Predictive models
KW  - Time series analysis
KW  - Supervised deep neural
KW  - networks
KW  - Car safety management
Y1  - 2021
U6  - https://doi.org/10.1007/s10618-021-00761-9
SN  - 1384-5810
SN  - 1573-756X
VL  - 35
IS  - 4
SP  - 1688
EP  - 1709
PB  - Springer
CY  - Dordrecht
ER  - 
TY  - JOUR
A1  - Zhelayskaya, Irina S.
A1  - Vasile, Ruggero
A1  - Shprits, Yuri Y.
A1  - Stolle, Claudia
A1  - Matzka, Jürgen
T1  - Systematic Analysis of Machine Learning and Feature Selection Techniques for Prediction of the Kp Index
JF  - Space Weather: The International Journal of Research and Applications
N2  - The Kp index is a measure of the midlatitude global geomagnetic activity and represents short-term magnetic variations driven by solar wind plasma and interplanetary magnetic field. The Kp index is one of the most widely used indicators for space weather alerts and serves as input to various models, such as for the thermosphere and the radiation belts. It is therefore crucial to predict the Kp index accurately. Previous work in this area has mostly employed artificial neural networks to nowcast Kp, based their inferences on the recent history of Kp and on solar wind measurements at L1. In this study, we systematically test how different machine learning techniques perform on the task of nowcasting and forecasting Kp for prediction horizons of up to 12 hr. Additionally, we investigate different methods of machine learning and information theory for selecting the optimal inputs to a predictive model. We illustrate how these methods can be applied to select the most important inputs to a predictive model of Kp and to significantly reduce input dimensionality. We compare our best performing models based on a reduced set of optimal inputs with the existing models of Kp, using different test intervals, and show how this selection can affect model performance.
KW  - Kp index
KW  - Predictive models
KW  - Feature selection
KW  - Machine learning
KW  - Validation
Y1  - 2019
U6  - https://doi.org/10.1029/2019SW002271
SN  - 1542-7390
VL  - 17
IS  - 10
SP  - 1461
EP  - 1486
PB  - American Geophysical Union
CY  - Washington
ER  - 
TY  - JOUR
A1  - Kayhan Wagner, Ezgi
A1  - Meyer, Marlene
A1  - O’Reilly, J.X.
A1  - Hunnius, Sabine
A1  - Bekkering, Harold
T1  - Nine-month-old infants update their predictive models of a changing environment
JF  - Developmental Cognitive Neuroscience : a journal for cognitive, affective and social developmental neuroscience
N2  - Humans generate internal models of their environment to predict events in the world. As the environments change, our brains adjust to these changes by updating their internal models. Here, we investigated whether and how 9-month-old infants differentially update their models to represent a dynamic environment. Infants observed a predictable sequence of stimuli, which were interrupted by two types of cues. Following the update cue, the pattern was altered, thus, infants were expected to update their predictions for the upcoming stimuli. Because the pattern remained the same after the no-update cue, no subsequent updating was required. Infants showed an amplified negative central (Nc) response when the predictable sequence was interrupted. Late components such as the PSW were also evoked in response to unexpected stimuli; however, we found no evidence for a differential response to the informational value of surprising cues at later stages of processing. Infants rather learned that surprising cues always signal a change in the environment that requires updating. Interestingly, infants responded with an amplified neural response to the absence of an expected change, suggesting a top-down modulation of early sensory processing in infants. Our findings corroborate emerging evidence showing that infants build predictive models early in life.
KW  - Internal models
KW  - Predictive models
KW  - Predictive processing
KW  - Development
KW  - Event-Related potentials
Y1  - 2019
U6  - https://doi.org/10.1016/j.dcn.2019.100680
SN  - 1878-9293
SN  - 1878-9307
VL  - 38
PB  - Elsevier
CY  - Amsterdam
ER  -