Blink: lightweight sample runs for cost optimization of big data applications. - In: New Trends in Database and Information Systems, (2022), S. 144-154
Distributed in-memory data processing engines accelerate iterative applications by caching datasets in memory rather than recomputing them in each iteration. Selecting a suitable cluster size for caching these datasets plays an essential role in achieving optimal performance. We present Blink, an autonomous sampling-based framework, which predicts sizes of cached datasets and selects optimal cluster size without relying on historical runs. We evaluate Blink on iterative, real-world, machine learning applications. With an average sample runs cost of 4.6% compared to the cost of optimal runs, Blink selects the optimal cluster size, saving up to 47.4% of execution cost compared to average cost.
https://doi.org/10.1007/978-3-031-15743-1_14
Cost modelling for optimal data placement in heterogeneous main memory. - In: Proceedings of the VLDB Endowment, ISSN 2150-8097, Bd. 15 (2022), 11, S. 2867-2880
The cost of DRAM contributes significantly to the operating costs of in-memory database management systems (IMDBMS). Persistent memory (PMEM) is an alternative type of byte-addressable memory that offers - in addition to persistence - higher capacities than DRAM at a lower price with the disadvantage of increased latencies and reduced bandwidth. This paper evaluates PMEM as a cheaper alternative to DRAM for storing table base data, which can make up a significant fraction of an IMDBMS' total memory footprint. Using a prototype implementation in the SAP HANA IMDBMS, we find that placing all table data in PMEM can reduce query performance in analytical benchmarks by more than a factor of two, while transactional workloads are less affected. To quantify the performance impact of placing individual data structures in PMEM, we propose a cost model based on a lightweight workload characterization. Using this model, we show how to place data pareto-optimally in the heterogeneous memory. Our evaluation demonstrates the accuracy of the model and shows that it is possible to place more than 75% of table data in PMEM while keeping performance within 10% of the DRAM baseline for two analytical benchmarks.
https://doi.org/10.14778/3551793.3551837
Error-driven chained multiple-subnetwork echo state network for time-series prediction. - In: IEEE sensors journal, ISSN 1558-1748, Bd. 22 (2022), 20, S. 19533-19542
Hybrid echo state networks (ESNs), a type of modified ESN, have been developed to improve the prediction accuracy of ESNs. However, they have been criticized for their computational complexity, which makes it difficult to use them directly in industrial applications. In this article, an error-driven chained multiple-subnetwork ESN (CESN) is proposed to build a simple structured hybrid network and improve its prediction accuracy. For this reason, a chain topology is generated to gradually reduce the residual error, while each subnetwork is trained separately. The weight matrix for each subnetwork does not need to be optimized, which reduces the computational cost. Meanwhile, the optimal number of subnetworks is determined on the basis of a given application. The efficiency of the proposed CESN is tested on a Santa Fe Laser and a public building dataset. Compared with ESN, 70% of the test data have been optimized by CESN for the public building dataset.
https://doi.org/10.1109/JSEN.2022.3200069
Brain electrical microstate features as biomarkers of a stable motor output. - In: Journal of neural engineering, ISSN 1741-2552, Bd. 19 (2022), 5, 056042, S. 1-16
Objective. The aim of the present study was to elucidate the brain dynamics underlying the maintenance of a constant force level exerted during a visually guided isometric contraction task by optimizing a predictive multivariate model based on global and spectral brain dynamics features. Approach. Electroencephalography (EEG) was acquired in 18 subjects who were asked to press a bulb and maintain a constant force level, indicated by a bar on a screen. For intervals of 500 ms, we calculated an index of force stability as well as indices of brain dynamics: microstate metrics (duration, occurrence, global explained variance, directional predominance) and EEG spectral amplitudes in the theta, low alpha, high alpha and beta bands. We optimized a multivariate regression model (partial least square (PLS)) where the microstate features and the spectral amplitudes were the input variables and the indexes of force stability were the output variables. The issues related to the collinearity among the input variables and to the generalizability of the model were addressed using PLS in a nested cross-validation approach. Main results. The optimized PLS regression model reached a good generalizability and succeeded to show the predictive value of microstates and spectral features in inferring the stability of the exerted force. Longer duration and higher occurrence of microstates, associated with visual and executive control networks, corresponded to better contraction performances, in agreement with the role played by the visual system and executive control network for visuo-motor integration. Significance. A combination of microstate metrics and brain rhythm amplitudes could be considered as biomarkers of a stable visually guided motor output not only at a group level, but also at an individual level. Our results may play an important role for a better understanding of the motor control in single trials or in real-time applications as well as in the study of motor control.
https://doi.org/10.1088/1741-2552/ac975b
The MORPHIA Project: first results of a long-term user study in an elderly care scenario from robotic point of view. - In: 54th International Symposium on Robotics, (2022), S. 66-73
In an aging society, efficiently organizing care taking tasks is of great importance including several players (here referred to as caregivers) like relatives, friends, professional caretakers, employees of retirement homes, clubs and so on. Especially for long-distance relationships, this can be burdensome and time-consuming. While supporting devices, like mobile phones or tablets, are slowly reaching the elder community, the drawbacks are obvious. These passive devices need to be handled by the elderly themselves, this includes an proper understanding of the operation, remembering to charge the devices, or even to hear incoming calls or messages. In the project MORPHIA, we target these drawbacks by combining a social communication platform on a tablet with a mobile robotic platform that can be remote-controlled by all mentioned actors of the supporting network or actively deliver messages emitted from the network. In this paper, we present the first stage of our demonstrator in terms of implemented hard- and software components. Since the price is a key factor for acceptance of such a system in the care community, we performed a technical assessment of these components based on our findings during the development process. In addition, we present the results of the first user tests with 5 participants over two weeks each between August and November 2021 (two further test iterations are planned for 2022 and 2023). This includes general usage of specific robotic services as well as technical benchmarks to assess the robustness of the developed system in domestic environments.
Effective re-identification of a multivariate process under model predictive control using information from plant-model mismatch detection. - In: Computer aided chemical engineering, ISSN 1570-7946, Bd. 49 (2022), S. 361-366
A process under model predictive control is required to be re-identified when plant-model mismatch (PMM) occurs. During data acquisition for re-identification, the process is excited to enable accurate re-identification. However, the excitation of the process worsens control performance. This research proposes a new method for re-identification that can deal with the problem. In the proposed method, only the inputs of the transfer functions that have significant PMM are excited, and, at the same time, the other inputs are manipulated to suppress the variations of the controlled variables. The usefulness of the proposed method was confirmed through a simulation case study of a 3-input, 3-output process. As a result, it was shown that the proposed method can reduce the mean absolute control error during data acquisition to 87% of that of an existing method without compromising model accuracy after re-identification.
https://doi.org/10.1016/B978-0-323-85159-6.50060-9
MVP-Net: multiple view pointwise semantic segmentation of large-scale point clouds. - In: Journal of WSCG, ISSN 1213-6964, Bd. 30 (2022), 1/2, S. 1-8
Semantic segmentation of 3D point cloud is an essential task for autonomous driving environment perception. The pipeline of most pointwise point cloud semantic segmentation methods includes points sampling, neighbor searching, feature aggregation, and classification. Neighbor searching method like K-nearest neighbors algorithm, KNN, has been widely applied. However, the complexity of KNN is always a bottleneck of efficiency. In this paper, we propose an end-to-end neural architecture, Multiple View Pointwise Net, MVP-Net, to efficiently and directly infer large-scale outdoor point cloud without KNN or any complex pre/postprocessing. Instead, assumption-based space filling curves and multi-rotation of point cloud methods are introduced to point feature aggregation and receptive field expanding. Numerical experiments show that the proposed MVP-Net is 11 times faster than the most efficient pointwise semantic segmentation method RandLA-Net [Qin20a] and achieves the same accuracy on the large-scale benchmark SemanticKITTI dataset.
https://www.doi.org/10.24132/JWSCG.2022.1
Comparison of fractional-order and integer-order H∞ control of a non-collocated two-mass oscillator. - In: IFAC-PapersOnLine, ISSN 2405-8963, Bd. 55 (2022), 25, S. 145-150
We consider the robust control of a two-mass oscillator with a dominant input delay. Our aim is to compare a fractional-order tuning approach including the partial compensation of non-minimum phase zeros with a classical H∞ loop-shaping design, since both these designs lead to a relatively high controller order. First of all a detailed physical model is derived and validated using measurement data. Based on the line arized model both controllers are designed to be comparable, i.e. they show a similar crossover frequency in the open loop and the final controller order is reduced to the same range for both designs. The major differences between both are the different methods how the feed-forward action is included. The loop-shaping approach with fractional-order elements relies on the plant inverse using a fat output, whereas the H∞ design incorporates a two-degree of freedom control, i.e. the reference signal is included into the known inputs of the generalized plant. Each controller is tested in simulation and experiment. As both open-loops are nearly identical in the frequency range of interest, the results from an input disturbance experiment show no major difference. The different design approaches of the feed forward path are clearly visible in the tracking experiment.
https://doi.org/10.1016/j.ifacol.2022.09.338
Fractional-order partial cancellation of integer-order poles and zeros. - In: IFAC-PapersOnLine, ISSN 2405-8963, Bd. 55 (2022), 25, S. 259-264
The key idea of this contribution is the partial compensation of non-minimum phase zeros or unstable poles. Therefore the integer-order zero/pole is split into a product of fractional-order pseudo zeros/poles. The amplitude and phase response of these fractional-order terms is derived to include these compensators into the loop-shaping design. Such compensators can be generalized to conjugate complex zeros/poles, and also implicit fractional-order terms can be applied. In the case of the non-minimum phase zero, its compensation leads to a higher phase margin and a steeper open-loop amplitude response around the crossover frequency resulting in a reduced undershooting in the step-response, as illustrated in the numerical example.
https://doi.org/10.1016/j.ifacol.2022.09.356
EVOLVE&hahog;INFOMAX: a new criterion for slow feature analysis of nonlinear dynamic system from an information-theoretical perspective. - In: IFAC-PapersOnLine, ISSN 2405-8963, Bd. 55 (2022), 20, S. 43-48
Slow feature analysis (SFA) has attracted much attention as a method for dynamic modelling. However, SFA has an inherent limitation in that it assumes that the dynamic behaviour is linear. In this paper, a new criterion for SFA in general dynamic systems is defined based on the motivation of maximising the information retained during system evolution, which is called EVOLVE&hahog;INFOMAX. The theoretical properties of this new criterion are rigorously justified, the optimisation function under EVOLVE&hahog;INFOMAX is proposed, and a tailored algorithm based on neural networks is designed. The case study on a simulated data set and the Tennessee Eastman process benchmark shows that the proposed method has better performance to extract slow features of nonlinear dynamical systems.
https://doi.org/10.1016/j.ifacol.2022.09.069