Search

As numerical simulations grow in size and complexity, they become increasingly resource-intensive in terms of time and energy. While specialized hardware accelerators often provide order-of-magnitude gains and are state of the art in other scientific fields, their availability and applicability in computational neuroscience is still limited. In this field, neuromorphic accelerators, particularly mixed-signal architectures like the BrainScaleS systems, offer the most significant performance benefits. These systems maintain a constant, accelerated emulation speed independent of network model and size. This is especially beneficial when traditional simulators reach their limits, such as when modeling complex neuron dynamics, incorporating plasticity mechanisms, or running long or repetitive experiments. However, the analog nature of these systems introduces new challenges. In this paper we demonstrate the capabilities and advantages of the BrainScaleS-1 system and how it can be used in combination with conventional software simulations. We report the emulation time and energy consumption for two biologically inspired networks adapted to the neuromorphic hardware substrate: a balanced random network based on Brunel and the cortical microcircuit from Potjans and Diesmann.

Arbor is a software library designed for efficient simulation of large-scale networks of biological neurons with detailed morphological structures. It combines customizable neuronal and synaptic mechanisms with high-performance computing, supporting multi-core CPU and GPU systems. In humans and other animals, synaptic plasticity processes play a vital role in cognitive functions, including learning and memory. Recent studies have shown that intracellular molecular processes in dendrites significantly influence single-neuron dynamics. However, for understanding how the complex interplay between dendrites and synaptic processes influences network dynamics, computational modeling is required. To enable the modeling of large-scale networks of morphologically detailed neurons with diverse plasticity processes, we have extended the Arbor library to the Plastic Arbor framework, supporting simulations of a large variety of spike-driven plasticity paradigms. To showcase the features of the new framework, we present examples of computational models, beginning with single-synapse dynamics, progressing to multi-synapse rules, and finally scaling up to large recurrent networks. While cross-validating our implementations by comparison with other simulators, we show that Arbor allows simulating plastic networks of multi-compartment neurons at nearly no additional cost in runtime compared to point-neuron simulations. Using the new framework, we have already been able to investigate the impact of dendritic structures on network dynamics across a timescale of several hours, showing a relation between the length of dendritic trees and the ability of the network to efficiently store information. By our extension of Arbor, we aim to provide a valuable tool that will support future studies on the impact of synaptic plasticity, especially, in conjunction with neuronal morphology, in large networks.

Quantum optimization is emerging as a prominent candidate for exploiting the capabilities of near-term quantum devices. Many application-relevant optimization tasks require the inclusion of inequality constraints, usually handled by enlarging the Hilbert space through the addition of slack variables. This approach, however, requires significant additional resources especially when considering multiple constraints. Here, we study an alternative direct implementation of these constraints within the QAOA algorithm, achieved using qudit-SUM gates, and compare it to the slack variable method generalized to qudits. We benchmark these approaches on three paradigmatic optimization problems. We find that the direct implementation of the inequality penalties vastly outperforms the slack variables method, especially when studying real-world inspired problems with many constraints. Within the direct penalty implementation, a linear energy penalty for unfeasible states outperforms other investigated functional forms, such as the canonical quadratic penalty. The proposed approach may thus be an enabling step for approaching realistic industry-scale and fundamental science problems with large numbers of inequality constraints.

Entropy scaling is a powerful technique that has been used successfully for predicting transport properties of pure components over a wide range of states. However, modeling mixture diffusion coefficients by entropy scaling is an unresolved task. We have tackled this issue and present an entropy scaling framework for predicting mixture self-diffusion as well as mutual diffusion in a thermodynamically consistent way. The predictions of the mixture diffusion coefficients are made based on information on the self-diffusion coefficients of the pure components and the infinite dilution diffusion coefficients. This is accomplished using information on the entropy of the mixture, which is taken here from molecular-based equations of state. The framework is flexible and can be applied together with any equation of state. Examples for the application of the new entropy scaling framework for the prediction of diffusion coefficients in mixtures illustrate its impressive performance. It enables predictions a wide range of temperatures and pressures including gaseous, liquid, supercritical, and metastable states - also for strongly non-ideal mixtures.

Solving combinatorial optimization problems on near-term quantum devices has gained a lot of attraction in recent years. Currently, most works have focused on single-objective problems, whereas many real-world applications need to consider multiple, mostly conflicting objectives, such as cost and quality. We present a variational quantum optimization algorithm to solve discrete multi-objective optimization problems on quantum computers. The proposed quantum multi-objective optimization (QMOO) algorithm incorporates all cost Hamiltonians representing the classical objective functions in the quantum circuit and produces a quantum state consisting of Pareto-optimal solutions in superposition. From this state we retrieve a set of solutions and utilize the widely applied hypervolume indicator to determine its quality as an approximation to the Pareto-front. The variational parameters of the QMOO circuit are tuned by maximizing the hypervolume indicator in a quantum-classical hybrid fashion. We show the effectiveness of the proposed algorithm on several benchmark problems with up to five objectives. We investigate the influence of the classical optimizer, the circuit depth and compare to results from classical optimization algorithms. We find that the algorithm is robust to shot noise and produces good results with as low as 128 measurement shots in each iteration. These promising result open the perspective to run the algorithm on near-term quantum hardware., Comment: revision of text, added shot noise analysis

We suggest a hybrid quantum-classical routine for the NP-hard Electric Vehicle Fleet Charging and Allocation Problem. The original formulation is a Mixed Integer Linear Program with continuous variables and inequality constraints. To separate inequality constraints that are difficult for quantum routines we use a decomposition in master and pricing problems: the former targets the assignment of vehicles to reservations and the latter suggests vehicle exploitation plans that respect the battery state-of-charge constraints. The master problem is equivalent to the search for an optimal set partition. In our hybrid scheme, the master problem is reformulated in a quadratic unconstrained binary optimization problem which can be solved with quantum annealing on the DWave Advantage system. On large instances, we benchmark the performance of the decomposition technique with classical and quantum-inspired metaheuristics: simulated annealing, tabu search, and vector annealing by NEC. The numerical results with purely classical solvers are comparable to the solutions from the traditional mixed integer linear programming approaches in terms of solution quality while being faster. In addition, it scales better to larger instances. The major advantage of the proposed approach is that it enables quantum-based methods for this realistic problem with many inequality constraints. We show this by initial studies on DWave hardware where optimal solutions can be found for small instances., Comment: 6 pages, 4 figures

Managing the response to natural disasters effectively can considerably mitigate their devastating impact. This work explores the potential of using supervised hybrid quantum machine learning to optimize emergency evacuation plans for cars during natural disasters. The study focuses on earthquake emergencies and models the problem as a dynamic computational graph where an earthquake damages an area of a city. The residents seek to evacuate the city by reaching the exit points where traffic congestion occurs. The situation is modeled as a shortest-path problem on an uncertain and dynamically evolving map. We propose a novel hybrid supervised learning approach and test it on hypothetical situations on a concrete city graph. This approach uses a novel quantum feature-wise linear modulation (FiLM) neural network parallel to a classical FiLM network to imitate Dijkstra's node-wise shortest path algorithm on a deterministic dynamic graph. Adding the quantum neural network in parallel increases the overall model's expressivity by splitting the dataset's harmonic and non-harmonic features between the quantum and classical components. The hybrid supervised learning agent is trained on a dataset of Dijkstra's shortest paths and can successfully learn the navigation task. The hybrid quantum network improves over the purely classical supervised learning approach by 7% in accuracy. We show that the quantum part has a significant contribution of 45.(3)% to the prediction and that the network could be executed on an ion-based quantum computer. The results demonstrate the potential of supervised hybrid quantum machine learning in improving emergency evacuation planning during natural disasters., Comment: 15 pages, 9 figures, 2 tables

Implementing resource efficient energy management systems in facilities and buildings becomes increasingly important in the transformation to a sustainable society. However, selecting a suitable configuration based on multiple, typically conflicting objectives, such as cost, robustness with respect to uncertainty of grid operation, or renewable energy utilization, is a difficult multi-criteria decision making problem. The recently developed concept identification technique can facilitate a decision maker by sorting configuration options into semantically meaningful groups (concepts). In this process, the partitioning of the objectives and design parameters into different sets (called description spaces) is a very important step. In this study we focus on utilizing the concept identification technique for finding relevant and viable energy management configurations from a very large data set of Pareto-optimal solutions. The data set consists of 20000 realistic Pareto-optimal building energy management configurations generated by a many-objective evolutionary optimization of a high quality Digital Twin energy management simulator. We analyze how the choice of description spaces, i.e., the partitioning of the objectives and parameters, impacts the type of information that can be extracted. We show that the decision maker can introduce constraints and biases into that process to meet expectations and preferences. The iterative approach presented in this work allows for the generation of valuable insights into trade-offs between specific objectives, and constitutes a powerful and flexible tool to support the decision making process when designing large and complex energy management systems., Comment: 18 pages, 8 figures, accepted at Energy Conversion and Management: X

The BrainScaleS-2 (BSS-2) system implements physical models of neurons as well as synapses and aims for an energy-efficient and fast emulation of biological neurons. When replicating neuroscientific experiments on BSS-2, a major challenge is finding suitable model parameters. This study investigates the suitability of the sequential neural posterior estimation (SNPE) algorithm for parameterizing a multi-compartmental neuron model emulated on the BSS-2 analog neuromorphic system. The SNPE algorithm belongs to the class of simulation-based inference methods and estimates the posterior distribution of the model parameters; access to the posterior allows quantifying the confidence in parameter estimations and unveiling correlation between model parameters. For our multi-compartmental model, we show that the approximated posterior agrees with experimental observations and that the identified correlation between parameters fits theoretical expectations. Furthermore, as already shown for software simulations, the algorithm can deal with high-dimensional observations and parameter spaces when the data is generated by emulations on BSS-2. These results suggest that the SNPE algorithm is a promising approach for automating the parameterization and the analyzation of complex models, especially when dealing with characteristic properties of analog neuromorphic substrates, such as trial-to-trial variations or limited parameter ranges.

The first-generation of BrainScaleS, also referred to as BrainScaleS-1, is a neuromorphic system for emulating large-scale networks of spiking neurons. Following a "physical modeling" principle, its VLSI circuits are designed to emulate the dynamics of biological examples: analog circuits implement neurons and synapses with time constants that arise from their electronic components' intrinsic properties. It operates in continuous time, with dynamics typically matching an acceleration factor of 10000 compared to the biological regime. A fault-tolerant design allows it to achieve wafer-scale integration despite unavoidable analog variability and component failures. In this paper, we present the commissioning process of a BrainScaleS-1 wafer module, providing a short description of the system's physical components, illustrating the steps taken during its assembly and the measures taken to operate it. Furthermore, we reflect on the system's development process and the lessons learned to conclude with a demonstration of its functionality by emulating a wafer-scale synchronous firing chain, the largest spiking network emulation ran with analog components and individual synapses to date.

Reducing the dimensions of ferroelectric materials down to the nanoscale has strong implications on the ferroelectric polarization pattern and on the ability to switch the polarization. As the size of ferroelectric domains shrinks to nanometer scale, the heterogeneity of the polarization pattern becomes increasingly pronounced, enabling a large variety of possible polar textures in nanocrystalline and nanocomposite materials. Critical to the understanding of fundamental physics of such materials and hence their applications in electronic nanodevices, is the ability to investigate their ferroelectric polarization at the nanoscale in a non-destructive way. We show that contact Kelvin probe force microscopy (cKPFM) combined with a k-means response clustering algorithm enables to measure the ferroelectric response at a mapping resolution of 8 nm. In a BaTiO3 thin film on silicon composed of tetragonal and hexagonal nanocrystals, we determine a nanoscale lateral distribution of discrete ferroelectric response clusters, fully consistent with the nanostructure determined by transmission electron microscopy. Moreover, we apply this data clustering method to the cKPFM responses measured at different temperatures, which allows us to follow the corresponding change in polarization pattern as the Curie temperature is approached and across the phase transition. This work opens up perspectives for mapping complex ferroelectric polarization textures such as curled/swirled polar textures that can be stabilized in epitaxial heterostructures and more generally mapping the polar domain distribution of any spatially-highly-heterogeneous ferroelectric materials.

Quantum two-level systems, i.e. qubits, form the basis for most quantum machine learning approaches that have been proposed throughout the years. However, higher dimensional quantum systems constitute a promising alternative and are increasingly explored in theory and practice. Here, we explore the capabilities of multi-level quantum systems, so-called qudits, for their use in a quantum machine learning context. We formulate classification and regression problems with the data re-uploading approach and demonstrate that a quantum circuit operating on a single qudit is able to successfully learn highly non-linear decision boundaries of classification problems such as the MNIST digit recognition problem. We demonstrate that the performance strongly depends on the relation between the qudit states representing the labels and the structure of labels in the training data set. Such a bias can lead to substantial performance improvement over qubit-based circuits in cases where the labels, the qudit states and the operators employed to encode the data are well-aligned. Furthermore, we elucidate the influence of the choice of the elementary operators and show that a squeezing operator is necessary to achieve good performances. We also show that there exists a trade-off for qudit systems between the number of circuit-generating operators in each processing layer and the total number of layers needed to achieve a given accuracy. Finally, we compare classification results from numerically exact simulations and their equivalent implementation on actual IBM quantum hardware. The findings of our work support the notion that qudit-based algorithms exhibit attractive traits and constitute a promising route to increasing the computational capabilities of quantum machine learning approaches.

Identifying meaningful concepts in large data sets can provide valuable insights into engineering design problems. Concept identification aims at identifying non-overlapping groups of design instances that are similar in a joint space of all features, but which are also similar when considering only subsets of features. These subsets usually comprise features that characterize a design with respect to one specific context, for example, constructive design parameters, performance values, or operation modes. It is desirable to evaluate the quality of design concepts by considering several of these feature subsets in isolation. In particular, meaningful concepts should not only identify dense, well separated groups of data instances, but also provide non-overlapping groups of data that persist when considering pre-defined feature subsets separately. In this work, we propose to view concept identification as a special form of clustering algorithm with a broad range of potential applications beyond engineering design. To illustrate the differences between concept identification and classical clustering algorithms, we apply a recently proposed concept identification algorithm to two synthetic data sets and show the differences in identified solutions. In addition, we introduce the mutual information measure as a metric to evaluate whether solutions return consistent clusters across relevant subsets. To support the novel understanding of concept identification, we consider a simulated data set from a decision-making problem in the energy management domain and show that the identified clusters are more interpretable with respect to relevant feature subsets than clusters found by common clustering algorithms and are thus more suitable to support a decision maker., Comment: 10 pages, 6 figures, published in proceedings of 2022 IEEE International Conference on Data Mining Workshops (ICDMW)

Multi-objective optimization problems whose objectives have different evaluation costs are commonly seen in the real world. Such problems are now known as multi-objective optimization problems with heterogeneous objectives (HE-MOPs). So far, however, only a few studies have been reported to address HE-MOPs, and most of them focus on bi-objective problems with one fast objective and one slow objective. In this work, we aim to deal with HE-MOPs having more than two black-box and heterogeneous objectives. To this end, we develop a multi-objective Bayesian evolutionary optimization approach to HE-MOPs by exploiting the different data sets on the cheap and expensive objectives in HE-MOPs to alleviate the search bias caused by the heterogeneous evaluation costs for evaluating different objectives. To make the best use of two different training data sets, one with solutions evaluated on all objectives and the other with those only evaluated on the fast objectives, two separate Gaussian process models are constructed. In addition, a new acquisition function that mitigates search bias towards the fast objectives is suggested, thereby achieving a balance between convergence and diversity. We demonstrate the effectiveness of the proposed algorithm by testing it on widely used multi-/many-objective benchmark problems whose objectives are assumed to be heterogeneously expensive.

Finding meaningful concepts in engineering application datasets which allow for a sensible grouping of designs is very helpful in many contexts. It allows for determining different groups of designs with similar properties and provides useful knowledge in the engineering decision making process. Also, it opens the route for further refinements of specific design candidates which exhibit certain characteristic features. In this work, an approach to define meaningful and consistent concepts in an existing engineering dataset is presented. The designs in the dataset are characterized by a multitude of features such as design parameters, geometrical properties or performance values of the design for various boundary conditions. In the proposed approach the complete feature set is partitioned into several subsets called description spaces. The definition of the concepts respects this partitioning which leads to several desired properties of the identified concepts. This cannot be achieved with state-of-the-art clustering or concept identification approaches. A novel concept quality measure is proposed, which provides an objective value for a given definition of concepts in a dataset. The usefulness of the measure is demonstrated by considering a realistic engineering dataset consisting of about 2500 airfoil profiles, for which the performance values (lift and drag) for three different operating conditions were obtained by a computational fluid dynamics simulation. A numerical optimization procedure is employed, which maximizes the concept quality measure and finds meaningful concepts for different setups of the description spaces, while also incorporating user preference. It is demonstrated how these concepts can be used to select archetypal representatives of the dataset which exhibit characteristic features of each concept., Comment: 19 pages, 14 figures, accepted at Advanced Engineering Informatics

Bayesian optimization has emerged at the forefront of expensive black-box optimization due to its data efficiency. Recent years have witnessed a proliferation of studies on the development of new Bayesian optimization algorithms and their applications. Hence, this paper attempts to provide a comprehensive and updated survey of recent advances in Bayesian optimization and identify interesting open problems. We categorize the existing work on Bayesian optimization into nine main groups according to the motivations and focus of the proposed algorithms. For each category, we present the main advances with respect to the construction of surrogate models and adaptation of the acquisition functions. Finally, we discuss the open questions and suggest promising future research directions, in particular with regard to heterogeneity, privacy preservation, and fairness in distributed and federated optimization systems.

Data stream classification is an important problem in the field of machine learning. Due to the non-stationary nature of the data where the underlying distribution changes over time (concept drift), the model needs to continuously adapt to new data statistics. Stream-based Active Learning (AL) approaches address this problem by interactively querying a human expert to provide new data labels for the most recent samples, within a limited budget. Existing AL strategies assume that labels are immediately available, while in a real-world scenario the expert requires time to provide a queried label (verification latency), and by the time the requested labels arrive they may not be relevant anymore. In this article, we investigate the influence of finite, time-variable, and unknown verification delay, in the presence of concept drift on AL approaches. We propose PRopagate (PR), a latency independent utility estimator which also predicts the requested, but not yet known, labels. Furthermore, we propose a drift-dependent dynamic budget strategy, which uses a variable distribution of the labelling budget over time, after a detected drift. Thorough experimental evaluation, with both synthetic and real-world non-stationary datasets, and different settings of verification latency and budget are conducted and analyzed. We empirically show that the proposed method consistently outperforms the state-of-the-art. Additionally, we demonstrate that with variable budget allocation in time, it is possible to boost the performance of AL strategies, without increasing the overall labeling budget., Comment: ENNS Best Paper Award at ICANN 2022

A frequent starting point of quantum computation platforms are two-state quantum systems, i.e., qubits. However, in the context of integer optimization problems, relevant to scheduling optimization and operations research, it is often more resource-efficient to employ quantum systems with more than two basis states, so-called qudits. Here, we discuss the quantum approximate optimization algorithm (QAOA) for qudit systems. We illustrate how the QAOA can be used to formulate a variety of integer optimization problems such as graph coloring problems or electric vehicle (EV) charging optimization. In addition, we comment on the implementation of constraints and describe three methods to include these into a quantum circuit of a QAOA by penalty contributions to the cost Hamiltonian, conditional gates using ancilla qubits, and a dynamical decoupling strategy. Finally, as a showcase of qudit-based QAOA, we present numerical results for a charging optimization problem mapped onto a max-$k$-graph coloring problem. Our work illustrates the flexibility of qudit systems to solve integer optimization problems., Comment: updated text and figures

Neuromorphic systems open up opportunities to enlarge the explorative space for computational research. However, it is often challenging to unite efficiency and usability. This work presents the software aspects of this endeavor for the BrainScaleS-2 system, a hybrid accelerated neuromorphic hardware architecture based on physical modeling. We introduce key aspects of the BrainScaleS-2 Operating System: experiment workflow, API layering, software design, and platform operation. We present use cases to discuss and derive requirements for the software and showcase the implementation. The focus lies on novel system and software features such as multi-compartmental neurons, fast re-configuration for hardware-in-the-loop training, applications for the embedded processors, the non-spiking operation mode, interactive platform access, and sustainable hardware/software co-development. Finally, we discuss further developments in terms of hardware scale-up, system usability and efficiency.

Quantum annealing is a heuristic quantum optimization algorithm that can be used to solve combinatorial optimization problems. In recent years, advances in quantum technologies have enabled the development of small- and intermediate-scale quantum processors that implement the quantum annealing algorithm for programmable use. Specifically, quantum annealing processors produced by D-Wave Systems have been studied and tested extensively in both research and industrial settings across different disciplines. In this paper we provide a literature review of the theoretical motivations for quantum annealing as a heuristic quantum optimization algorithm, the software and hardware that is required to use such quantum processors, and the state-of-the-art applications and proofs-of-concepts that have been demonstrated using them. The goal of our review is to provide a centralized and condensed source regarding applications of quantum annealing technology. We identify the advantages, limitations, and potential of quantum annealing for both researchers and practitioners from various fields., Comment: major revision with extended noise section and discussion of alternative platforms

An important issue during an engineering design process is to develop an understanding which design parameters have the most influence on the performance. Especially in the context of optimization approaches this knowledge is crucial in order to realize an efficient design process and achieve high-performing results. Information theory provides powerful tools to investigate these relationships because measures are model-free and thus also capture non-linear relationships, while requiring only minimal assumptions on the input data. We therefore propose to use recently introduced information-theoretic methods and estimation algorithms to find the most influential input parameters in optimization results. The proposed methods are in particular able to account for interactions between parameters, which are often neglected but may lead to redundant or synergistic contributions of multiple parameters. We demonstrate the application of these methods on optimization data from aerospace engineering, where we first identify the most relevant optimization parameters using a recently introduced information-theoretic feature-selection algorithm that accounts for interactions between parameters. Second, we use the novel partial information decomposition (PID) framework that allows to quantify redundant and synergistic contributions between selected parameters with respect to the optimization outcome to identify parameter interactions. We thus demonstrate the power of novel information-theoretic approaches in identifying relevant parameters in optimization runs and highlight how these methods avoid the selection of redundant parameters, while detecting interactions that result in synergistic contributions of multiple parameters., Comment: Preprint. Accepted at SSCI 2021. This work has been submitted to the IEEE for possible publication

Most existing multiobjetive evolutionary algorithms (MOEAs) implicitly assume that each objective function can be evaluated within the same period of time. Typically. this is untenable in many real-world optimization scenarios where evaluation of different objectives involves different computer simulations or physical experiments with distinct time complexity. To address this issue, a transfer learning scheme based on surrogate-assisted evolutionary algorithms (SAEAs) is proposed, in which a co-surrogate is adopted to model the functional relationship between the fast and slow objective functions and a transferable instance selection method is introduced to acquire useful knowledge from the search process of the fast objective. Our experimental results on DTLZ and UF test suites demonstrate that the proposed algorithm is competitive for solving bi-objective optimization where objectives have non-uniform evaluation times.

Detecting drifts in data is essential for machine learning applications, as changes in the statistics of processed data typically has a profound influence on the performance of trained models. Most of the available drift detection methods are either supervised and require access to the true labels during inference time, or they are completely unsupervised and aim for changes in distributions without taking label information into account. We propose a novel task-sensitive semi-supervised drift detection scheme, which utilizes label information while training the initial model, but takes into account that supervised label information is no longer available when using the model during inference. It utilizes a constrained low-dimensional embedding representation of the input data. This way, it is best suited for the classification task. It is able to detect real drift, where the drift affects the classification performance, while it properly ignores virtual drift, where the classification performance is not affected by the drift. In the proposed framework, the actual method to detect a change in the statistics of incoming data samples can be chosen freely. Experimental evaluation on nine benchmarks datasets, with different types of drift, demonstrates that the proposed framework can reliably detect drifts, and outperforms state-of-the-art unsupervised drift detection approaches., Comment: Preprint. Submitted at SSCI 2021. This work has been submitted to the IEEE for possible publication

Selecting a minimal feature set that is maximally informative about a target variable is a central task in machine learning and statistics. Information theory provides a powerful framework for formulating feature selection algorithms -- yet, a rigorous, information-theoretic definition of feature relevancy, which accounts for feature interactions such as redundant and synergistic contributions, is still missing. We argue that this lack is inherent to classical information theory which does not provide measures to decompose the information a set of variables provides about a target into unique, redundant, and synergistic contributions. Such a decomposition has been introduced only recently by the partial information decomposition (PID) framework. Using PID, we clarify why feature selection is a conceptually difficult problem when approached using information theory and provide a novel definition of feature relevancy and redundancy in PID terms. From this definition, we show that the conditional mutual information (CMI) maximizes relevancy while minimizing redundancy and propose an iterative, CMI-based algorithm for practical feature selection. We demonstrate the power of our CMI-based algorithm in comparison to the unconditional mutual information on benchmark examples and provide corresponding PID estimates to highlight how PID allows to quantify information contribution of features and their interactions in feature-selection problems., Comment: 44 pages, 12 figures. Reorganization and shortening of manuscript, added Appendix with theoretical guarantees, background information on the algorithm used, and an additional example application on a larger problem. Minor text editing

In complex industrial settings, it is common practice to monitor the operation of machines in order to detect undesired states, adjust maintenance schedules, optimize system performance or collect usage statistics of individual machines. In this work, we focus on estimating the power output of a Combined Heat and Power (CHP) machine of a medium-sized company facility by analyzing the total facility power consumption. We formulate the problem as a time-series classification problem where the class label represents the CHP power output. As the facility is fully instrumented and sensor measurements from the CHP are available, we generate the training labels in an automated fashion from the CHP sensor readings. However, sensor failures result in mislabeled training data samples which are hard to detect and remove from the dataset. Therefore, we propose a novel multi-task deep learning approach that jointly trains a classifier and an autoencoder with a shared embedding representation. The proposed approach targets to gradually correct the mislabelled data samples during training in a self-supervised fashion, without any prior assumption on the amount of label noise. We benchmark our approach on several time-series classification datasets and find it to be comparable and sometimes better than state-of-the-art methods. On the real-world use-case of predicting the CHP power output, we thoroughly evaluate the architectural design choices and show that the final architecture considerably increases the robustness of the learning process and consistently beats other recent state-of-the-art algorithms in the presence of unstructured as well as structured label noise., Comment: Accepted in Applied Data Science track at ECML 2021

Convolutional neural network training can suffer from diverse issues like exploding or vanishing gradients, scaling-based weight space symmetry and covariant-shift. In order to address these issues, researchers develop weight regularization methods and activation normalization methods. In this work we propose a weight soft-regularization method based on the Oblique manifold. The proposed method uses a loss function which pushes each weight vector to have a norm close to one, i.e. the weight matrix is smoothly steered toward the so-called Oblique manifold. We evaluate our method on the very popular CIFAR-10, CIFAR-100 and ImageNet 2012 datasets using two state-of-the-art architectures, namely the ResNet and wide-ResNet. Our method introduces negligible computational overhead and the results show that it is competitive to the state-of-the-art and in some cases superior to it. Additionally, the results are less sensitive to hyperparameter settings such as batch size and regularization factor.

Computational Fluid Dynamics (CFD) simulations are a very important tool for many industrial applications, such as aerodynamic optimization of engineering designs like cars shapes, airplanes parts etc. The output of such simulations, in particular the calculated flow fields, are usually very complex and hard to interpret for realistic three-dimensional real-world applications, especially if time-dependent simulations are investigated. Automated data analysis methods are warranted but a non-trivial obstacle is given by the very large dimensionality of the data. A flow field typically consists of six measurement values for each point of the computational grid in 3D space and time (velocity vector values, turbulent kinetic energy, pressure and viscosity). In this paper we address the task of extracting meaningful results in an automated manner from such high dimensional data sets. We propose deep learning methods which are capable of processing such data and which can be trained to solve relevant tasks on simulation data, i.e. predicting drag and lift forces applied on an airfoil. We also propose an adaptation of the classical hand crafted features known from computer vision to address the same problem and compare a large variety of descriptors and detectors. Finally, we compile a large dataset of 2D simulations of the flow field around airfoils which contains 16000 flow fields with which we tested and compared approaches. Our results show that the deep learning-based methods, as well as hand crafted feature based approaches, are well-capable to accurately describe the content of the CFD simulation output on the proposed dataset.

In the context of optimization approaches to engineering applications, time-consuming simulations are often utilized which can be configured to deliver solutions for various levels of accuracy, commonly referred to as different fidelity levels. It is common practice to train hierarchical surrogate models on the objective functions in order to speed-up the optimization process. These operate under the assumption that there is a correlation between the high- and low-fidelity versions of the problem that can be exploited to cheaply gain information. In the practical scenario where the computational budget has to be allocated between multiple fidelities, limited guidelines are available to help make that division. In this paper we evaluate a range of different choices for a two-fidelity setup that provide helpful intuitions about the trade-off between evaluating in high- or low-fidelity. We present a heuristic method based on subsampling from an initial Design of Experiments (DoE) to find a suitable division of the computational budget between the fidelity levels. This enables the setup of multi-fidelity optimizations which utilize the available computational budget efficiently, independent of the multi-fidelity model used., Comment: 12 pages, 9 figures. This is an original manuscript of an article published by Taylor & Francis in Engineering Optimization on 2022-05-16, available online: http://www.tandfonline.com/10.1080/0305215X.2022.2052286

The continuously growing amount of monitored data in the Industry 4.0 context requires strong and reliable anomaly detection techniques. The advancement of Digital Twin technologies allows for realistic simulations of complex machinery, therefore, it is ideally suited to generate synthetic datasets for the use in anomaly detection approaches when compared to actual measurement data. In this paper, we present novel weakly-supervised approaches to anomaly detection for industrial settings. The approaches make use of a Digital Twin to generate a training dataset which simulates the normal operation of the machinery, along with a small set of labeled anomalous measurement from the real machinery. In particular, we introduce a clustering-based approach, called Cluster Centers (CC), and a neural architecture based on the Siamese Autoencoders (SAE), which are tailored for weakly-supervised settings with very few labeled data samples. The performance of the proposed methods is compared against various state-of-the-art anomaly detection algorithms on an application to a real-world dataset from a facility monitoring system, by using a multitude of performance measures. Also, the influence of hyper-parameters related to feature extraction and network architecture is investigated. We find that the proposed SAE based solutions outperform state-of-the-art anomaly detection approaches very robustly for many different hyper-parameter settings on all performance measures., Comment: in IEEE Transactions on Industrial Informatics

We present software facilitating the usage of the BrainScaleS-2 analog neuromorphic hardware system as an inference accelerator for artificial neural networks. The accelerator hardware is transparently integrated into the PyTorch machine learning framework using its extension interface. In particular, we provide accelerator support for vector-matrix multiplications and convolutions; corresponding software-based autograd functionality is provided for hardware-in-the-loop training. Automatic partitioning of neural networks onto one or multiple accelerator chips is supported. We analyze implementation runtime overhead during training as well as inference, provide measurements for existing setups and evaluate the results in terms of the accelerator hardware design limitations. As an application of the introduced framework, we present a model that classifies activities of daily living with smartphone sensor data.

BrainScaleS-1 is a wafer-scale mixed-signal accelerated neuromorphic system targeted for research in the fields of computational neuroscience and beyond-von-Neumann computing. The BrainScaleS Operating System (BrainScaleS OS) is a software stack giving users the possibility to emulate networks described in the high-level network description language PyNN with minimal knowledge of the system. At the same time, expert usage is facilitated by allowing to hook into the system at any depth of the stack. We present operation and development methodologies implemented for the BrainScaleS-1 neuromorphic architecture and walk through the individual components of BrainScaleS OS constituting the software stack for BrainScaleS-1 platform operation.

Aerodynamic shape optimization has established itself as a valuable tool in the engineering design process to achieve highly efficient results. A central aspect for such approaches is the mapping from the design parameters which encode the geometry of the shape to be improved to the quality criteria which describe its performance. The choices to be made in the setup of the optimization process strongly influence this mapping and thus are expected to have a profound influence on the achievable result. In this work we explore the influence of such choices on the effects on the shape optimization of a turbofan rotor blade as it can be realized within an aircraft engine design process. The blade quality is assessed by realistic three dimensional computational fluid dynamics (CFD) simulations. We investigate the outcomes of several optimization runs which differ in various configuration options, such as optimization algorithm, initialization, number of degrees of freedom for the parametrization. For all such variations, we generally find that the achievable improvement of the blade quality is comparable for most settings and thus rather insensitive to the details of the setup. On the other hand, even supposedly minor changes in the settings, such as using a different random seed for the initialization of the optimizer algorithm, lead to very different shapes. Optimized shapes which show comparable performance usually differ quite strongly in their geometries over the complete blade. Our analyses indicate that the fitness landscape for such a realistic turbofan rotor blade optimization is highly multi-modal with many local optima, where very different shapes show similar performance.

We investigate the trade-off between the robustness against random and targeted removal of nodes from a network. To this end we utilize the stochastic block model to study ensembles of infinitely large networks with arbitrary large-scale structures. We present results from numerical two-objective optimization simulations for networks with various fixed mean degree and number of blocks. The results provide strong evidence that three different blocks are sufficient to realize the best trade-off between the two measures of robustness, i.e.\ to obtain the complete front of Pareto-optimal networks. For all values of the mean degree, a characteristic three block structure emerges over large parts of the Pareto-optimal front. This structure can be often characterized as a core-periphery structure, composed of a group of core nodes with high degree connected among themselves and to a periphery of low-degree nodes, in addition to a third group of nodes which is disconnected from the periphery, and weakly connected to the core. Only at both extremes of the Pareto-optimal front, corresponding to maximal robustness against random and targeted node removal, a two-block core-periphery structure or a one-block fully random network are found, respectively.

The massively parallel nature of biological information processing plays an important role for its superiority to human-engineered computing devices. In particular, it may hold the key to overcoming the von Neumann bottleneck that limits contemporary computer architectures. Physical-model neuromorphic devices seek to replicate not only this inherent parallelism, but also aspects of its microscopic dynamics in analog circuits emulating neurons and synapses. However, these machines require network models that are not only adept at solving particular tasks, but that can also cope with the inherent imperfections of analog substrates. We present a spiking network model that performs Bayesian inference through sampling on the BrainScaleS neuromorphic platform, where we use it for generative and discriminative computations on visual data. By illustrating its functionality on this platform, we implicitly demonstrate its robustness to various substrate-specific distortive effects, as well as its accelerated capability for computation. These results showcase the advantages of brain-inspired physical computation and provide important building blocks for large-scale neuromorphic applications., Comment: This preprint has been published 2019 November 14. Please cite as: Kungl A. F. et al. (2019) Accelerated Physical Emulation of Bayesian Inference in Spiking Neural Networks. Front. Neurosci. 13:1201. doi: 10.3389/fnins.2019.01201

Despite being originally inspired by the central nervous system, artificial neural networks have diverged from their biological archetypes as they have been remodeled to fit particular tasks. In this paper, we review several possibilites to reverse map these architectures to biologically more realistic spiking networks with the aim of emulating them on fast, low-power neuromorphic hardware. Since many of these devices employ analog components, which cannot be perfectly controlled, finding ways to compensate for the resulting effects represents a key challenge. Here, we discuss three different strategies to address this problem: the addition of auxiliary network components for stabilizing activity, the utilization of inherently robust architectures and a training method for hardware-emulated networks that functions without perfect knowledge of the system's dynamics and parameters. For all three scenarios, we corroborate our theoretical considerations with experimental results on accelerated analog neuromorphic platforms., Comment: accepted at ISCAS 2017

Emulating spiking neural networks on analog neuromorphic hardware offers several advantages over simulating them on conventional computers, particularly in terms of speed and energy consumption. However, this usually comes at the cost of reduced control over the dynamics of the emulated networks. In this paper, we demonstrate how iterative training of a hardware-emulated network can compensate for anomalies induced by the analog substrate. We first convert a deep neural network trained in software to a spiking network on the BrainScaleS wafer-scale neuromorphic system, thereby enabling an acceleration factor of 10 000 compared to the biological time domain. This mapping is followed by the in-the-loop training, where in each training step, the network activity is first recorded in hardware and then used to compute the parameter updates in software via backpropagation. An essential finding is that the parameter updates do not have to be precise, but only need to approximately follow the correct gradient, which simplifies the computation of updates. Using this approach, after only several tens of iterations, the spiking network shows an accuracy close to the ideal software-emulated prototype. The presented techniques show that deep spiking networks emulated on analog neuromorphic devices can attain good computational performance despite the inherent variations of the analog substrate., Comment: 8 pages, 10 figures, submitted to IJCNN 2017

Hexagonally aligned, free-standing silicon nanowire (SiNW) arrays serve as photonic resonators which, as compared to a silicon (Si) thin film, do not only absorb more visible (VIS) and near-infrared (NIR) light, but also show an inherent photonic light concentration that enhances their performance as solar absorbers. Using numerical simulations we show, how light concentration is induced by high optical cross sections of the individual SiNWs but cannot be optimized independently of the SiNW array absorption. While an ideal spatial density exists, for which the SiNW array absorption for VIS and NIR wavelengths reaches a maximum, the spatial correlation of SiNWs in an array suppresses the formation of optical Mie modes responsible for light concentration. We show that different from SiNWs with straight sidewalls, arrays of inverted silicon nanocones (SiNCs) permit to avoid the mode suppression. In fact they give rise to an altered set of photonic modes which is induced by the spatial correlation of SiNCs in the array, and therefore show a higher degree of freedom to independently optimize light absorption and light concentration. Apart from explaining the good light absorbing and concentrating properties of SiNC arrays, the work justifies a revaluation of SiNW arrays as optical absorbers.

Silicon nanowires (SiNWs) attached to a wafer substrate are converted to inversely tapered silicon nanocones (SiNCs). After excitation with visible light, individual SiNCs show a 200-fold enhanced integral band-to-band luminescence as compared to a straight SiNW reference. Furthermore, the reverse taper is responsible for multifold emission peaks in addition to the relatively broad nearinfrared (NIR) luminescence spectrum. A thorough numerical mode analysis reveals that unlike a SiNW the inverted SiNC sustains a multitude of leaky whispering gallery modes. The modes are unique to this geometry and they are characterized by a relatively high quality factor ($Q = 1300$) and a low mode volume ($0.2 < ({\lambda}/n_{eff})^3 < 4$). In addition they show a vertical out coupling of the optically excited NIR luminescence with a numerical aperture as low as 0.22. Estimated Purcell factors $F_p \propto Q/V_m$ of these modes can explain the enhanced luminescence in individual emission peaks as compared to the SiNW reference. Investigating the relation between the SiNC geometry and the mode formation leads to simple design rules that permit to control the number and wavelength of the hosted modes and therefore the luminescent emission peaks.