The rising energy production costs and the increasing reliance on volatile renewable sources have driven the need for more efficient power system redispatch strategies. In this work, we re-interpret the redispatch problem as a multi-objective combinatorial optimization task within the Quadratic Unconstrained Binary Optimization (QUBO) framework, suitable for adiabatic quantum computing. Our contributions include a novel normalized unbalanced penalty method that integrates inequality constraints via a quadratic Taylor expansion and an$\alpha$-Expansion algorithm that allows us to address largescale redispatch instances and to integrate temporal adjacent state switching constraints directly into the algorithm. Our experiments are conducted on open data of the German power system. Our results, obtained via numerical simulation and from an actual D-Wave Advantage quantum annealer, validate the viability of our formulation and demonstrate that our algorithm scales to large problem instances.

Multi-Agent Path Finding (MAPF) focuses on determining conflict-free paths for multiple agents navigating through a shared space to reach specified goal locations. This problem becomes computationally challenging, particularly when handling large numbers of agents, as frequently encountered in practical applications like coordinating autonomous vehicles. Quantum Computing (QC) is a promising candidate in overcoming such limits. However, current quantum hardware is still in its infancy and thus limited in terms of computing power and error robustness. In this work, we present the first optimal hybrid quantum-classical MAPF algorithms which are based on branch-and-cut-and-prize. QC is integrated by iteratively solving QUBO problems, based on conflict graphs. Experiments on actual quantum hardware and results on benchmark data suggest that our approach dominates previous QUBO formulations and state-of-the-art MAPF solvers.

In this work we explore the use of quantum computers in solving the NP-hard placement problem of the Field-Programmable Gate Array (FPGA) implementation phase and introduce a novel approach suited for current quantum hardware sizes. Adiabatic quantum computing (AQC), with its capability to traverse expansive solution spaces, is a good fit for addressing this combinatorial problem with its exponentially large solution space. Instead of solving a single the whole problem at once, we re-formulate the placement problem as a series of so called quadratic unconstrained binary optimization (QUBO) problems which are subsequently solved via AQC. Our novel formulation facilitates a straight-forward integration of design constraints. Moreover, the size of the sub-problems can be conveniently adapted to the available hardware capabilities. Beside the sole proposal of a novel method, we ask whether contemporary quantum hardware is resilient enough to find placements for real-world-sized FPGAs. A numerical evaluation on a D-Wave Advantage 5.4 quantum annealer suggests that the answer is in the affirmative.

Abstract The ability to compute the exact divergence between two high-dimensional distributions is useful in many applications, but doing so naively is intractable. Computing the $$\alpha \beta $$ α β -divergence—a family of divergences that includes the Kullback–Leibler divergence and Hellinger distance—between the joint distribution of two decomposable models, i.e., chordal Markov networks, can be done in time exponential in the treewidth of these models. Extending this result, we propose an approach to compute the exact $$\alpha \beta $$ α β -divergence between any marginal or conditional distribution of two decomposable models. In order to do so tractably, we provide a decomposition over the marginal and conditional distributions of decomposable models. We then show how our method can be used to analyze distributional changes by first applying it to the benchmark image dataset QMNIST and a dataset containing observations from various areas at the Roosevelt Nation Forest and their cover type. Finally, based on our framework, we propose a novel way to quantify the error in contemporary superconducting quantum computers.

Abstract Recent advances in practical quantum computing have led to a variety of cloud-based quantum computing platforms that allow researchers to evaluate their algorithms on noisy intermediate-scale quantum devices. A common property of quantum computers is that they can exhibit instances of true randomness as opposed to pseudo-randomness obtained from classical systems. Investigating the effects of such true quantum randomness in the context of machine learning is appealing, and recent results vaguely suggest that benefits can indeed be achieved from the use of quantum random numbers. To shed some more light on this topic, we empirically study the effects of hardware-biased quantum random numbers on the initialization of artificial neural network weights in numerical experiments. We find no statistically significant difference in comparison with unbiased quantum random numbers as well as biased and unbiased random numbers from a classical pseudo-random number generator. The quantum random numbers for our experiments are obtained from real quantum hardware.

There are many applications that benefit from computing the exact divergence between 2 discrete probability measures, including machine learning. Unfortunately, in the absence of any assumptions on the structure or independencies within these distributions, computing the divergence between them is an intractable problem in high dimensions. We show that we are able to compute a wide family of functionals and divergences, such as the alpha-beta divergence, between two decomposable models, i.e. chordal Markov networks, in time exponential to the treewidth of these models. The alpha-beta divergence is a family of divergences that include popular divergences such as the Kullback-Leibler divergence, the Hellinger distance, and the chi-squared divergence. Thus, we can accurately compute the exact values of any of this broad class of divergences to the extent to which we can accurately model the two distributions using decomposable models.

The ability to compute the exact divergence between two high-dimensional distributions is useful in many applications but doing so naively is intractable. Computing the alpha-beta divergence—a family of divergences that includes the Kullback-Leibler divergence and Hellinger distance—between the joint distribution of two decomposable models, i.e chordal Markov networks, can be done in time exponential in the treewidth of these models. However, reducing the dissimilarity between two high-dimensional objects to a single scalar value can be uninformative. Furthermore, in applications such as supervised learning, the divergence over a conditional distribution might be of more interest. Therefore, we propose an approach to compute the exact alpha-beta divergence between any marginal or conditional distribution of two decomposable models. Doing so tractably is non-trivial as we need to decompose the divergence between these distributions and therefore, require a decomposition over the marginal and conditional distributions of these models. Consequently, we provide such a decomposition and also extend existing work to compute the marginal and conditional alpha-beta divergence between these decompositions. We then show how our method can be used to analyze distributional changes by first applying it to a benchmark image dataset. Finally, based on our framework, we propose a novel way to quantify the error in contemporary superconducting quantum computers. Code for all experiments is available at: https://lklee.dev/pub/2023-icdm/code

Variational quantum circuits build the foundation for various classes of quantum algorithms. In a nutshell, the weights of a parametrized quantum circuit are varied until the empirical sampling distribution of the circuit is sufficiently close to a desired outcome. Numerical first-order methods are applied frequently to fit the parameters of the circuit, but most of the time, the circuit itself, that is, the actual composition of gates, is fixed. Methods for optimizing the circuit design jointly with the weights have been proposed, but empirical results are rather scarce. Here, we consider a simple evolutionary strategy that addresses the trade-off between finding appropriate circuit ar-chitectures and parameter tuning. We evaluate our method both via simulation and on actual quantum hardware. Our benchmark problems include the transverse field Ising Hamiltonian and the Sherrington-Kirkpatrick spin model. Despite the shortcomings of current noisy intermediate-scale quantum hardware, we find only a minor slowdown on actual quantum machines compared to simulations. Moreover, we investigate which mutation operations most significantly contribute to the optimization. The results provide intuition on how randomized search heuristics behave on actual quantum hardware and layout a path for further refinement of evolutionary quantum gate circuits.

Probabilistic methods serve as the underlying frame-work of various machine learning techniques. When using these models, a central problem is that of computing the partition function, whose computation is intractable for many models of interest. Here, we present the first quantum-inspired method that is especially designed for computing fast approximations to the partition function. Our approach uses a novel hardware solver for quadratic unconstrained binary optimization problems that relies on evolutionary computation. The specialized design allows us to assess millions of candidate solutions per second, leading to high quality maximum a-posterior (MAP) estimates, even for hard instances. We investigate the expected run-time of our solver and devise new ultra-sparse parity constraints to combine our device with the WISH approximation scheme. A SIMD-like packing strategy further allows us to solve multiple MAP instances at once, resulting in high efficiency and an additional speed-up. Numerical experiments show that our quantum-inspired approach produces accurate and robust results. While pure software implementations of the WISH algorithm typically run on large compute clusters with hundreds of CPUs, our results are achieved on two FPGA boards which both consume below 10 Watts. Moreover, our results extend seamlessly to adiabatic quantum computers.

Probabilistic timing guarantees enable a tradeoff between system safety and hardware costs in embedded real-time systems. A key metric for assessing whether timing requirements can be satisfied with sufficiently high probability is the worst-case deadline failure probability (WCDFP). This paper studies the WCDFP under earliest-deadline first (EDF) scheduling for tasks with several probabilistic execution modes (e.g., a low-needs "typical" mode and a resource-intensive "exceptional" mode). Under EDF, no known approach can bound the WCDFP for practically sized workloads since the time complexity of prior approaches is exponential in the number of jobs.This paper examines the structure of the EDF WCDFP problem and establishes a safe, efficiently computable over-approximation by restricting the analysis to a set of specific intervals and providing a criterion to stop the derivation early without risking under-approximation. The analysis first assumes independent jobs and is then extended to handle dependencies (i.e., acyclic task chains). An evaluation shows that (i) even if 99.9999% of the jobs must meet their deadlines, a significantly higher utilization is possible than in the deterministic case, (ii) the analysis is scalable to 30 tasks with more than 1060jobs in the hyperperiod, and (iii) assuming independence in the presence of dependent tasks can severely under-estimate the WCDFP.

Spatio-temporal data sets such as satellite image series are of utmost importance for understanding global developments like climate change or urbanization. However, incompleteness of data can greatly impact usability and knowledge discovery. In fact, there are many cases where not a single data point in the set is fully observed. For filling gaps, we introduce a novel approach that utilizes Markov random fields (MRFs). We extend the probabilistic framework to also consider empirical prior information, which allows to train even on highly incomplete data. Moreover, we devise a way to make discrete MRFs predict continuous values via state superposition. Experiments on real-world remote sensing imagery suffering from cloud cover show that the proposed approach outperforms state-of-the-art gap filling techniques.

Exploiting big data knowledge on small devices will pave the way for building truly cognitive Internet of Things (IoT) systems. Although machine learning has led to great advancements for IoT-based data analytics, there remains a huge methodological gap for the deployment phase of trained machine learning models. For given resource-constrained platforms such as Microcontroller Units (MCUs), model choice and parametrization are typically performed based on heuristics or analytical models. However, these approaches are only able to provide rough estimates of the required system resources as they do not consider the interplay of hardware, compilerspecific optimizations, and code dependencies. In this paper, we present the novel open source framework LIghtweight Machine learning for IoT Systems (LIMITS), which applies a platform-in-the-loop approach explicitly considering the actual compilation toolchain of the target IoT platform. LIMITS focuses on high-level tasks such as experiment automation, platform-specific code generation, and sweet spot determination. The solid foundations of validated low-level model implementations are provided by the coupled well-established data analysis framework Waikato Environment for Knowledge Analysis (WEKA). We apply and validate LIMITS in two case studies focusing on cellular data rate prediction and radio-based vehicle classification, where we compare different learning models and real world IoT platforms with memory constraints from 16 kB to 4 MB and demonstrate its potential to catalyze the development of machine learning-enabled IoT systems.

Artificial intelligence for autonomous driving must meet strict requirements on safety and robustness, which motivates the thorough validation of learned models. However, current validation approaches mostly require ground truth data and are thus both cost-intensive and limited in their applicability. We propose to overcome these limitations by a model agnostic validation using a-priori knowledge from street maps. In particular, we show how to validate semantic segmentation masks and demonstrate the potential of our approach using OpenStreetMap. We introduce validation metrics that indicate false positive or negative road segments. Besides the validation approach, we present a method to correct the vehicle's GPS position so that a more accurate localization can be used for the street-map based validation. Lastly, we present quantitative results on the Cityscapes dataset indicating that our validation approach can indeed uncover errors in semantic segmentation masks.

Ubiquitously deployed Internet of Things (IoT)-based automatic vehicle classification systems will catalyze data-driven traffic flow optimization in future smart cities and will transform the road infrastructure itself into a dynamically sensing cyber-physical system. Although a wide range of different traffic sensing systems has been proposed, the existing solutions are not yet able to simultaneously satisfy the multitude of requirements, e.g., accuracy, robustness, cost efficiency, and privacy preservation. In this article, we present a novel approach, which exploits radio fingerprints-multidimensional attenuation patterns of wireless signals-for accurate and robust vehicle detection and classification. The proposed system can be deployed in a highly cost-efficient manner as it relies on off-the-shelf embedded devices which are installed into existing delineator posts. In a comprehensive field evaluation campaign, the performance of the radio fingerprinting-based approach is analyzed within an experimental live deployment on a German highway, where it is able to achieve a binary classification success ratio of more than 99% and an overall accuracy of 93.83% for a classification task with seven different classes.

The exploitation of vehicles as mobile sensors acts as a catalyst for novel crowdsensing-based applications such as intelligent traffic control and distributed weather forecast. However, the massive increases in Machine-type Communication (MTC) highly stress the capacities of the network infrastructure. With the system-immanent limitation of resources in cellular networks and the resource competition between human cell users and MTC, more resource-efficient channel access methods are required in order to improve the coexistence of the different communicating entities. In this paper, we present a machine learning-enabled transmission scheme for client-side opportunistic data transmission. By considering the measured channel state as well as the predicted future channel behavior, delay-tolerant MTC is performed with respect to the anticipated resource-efficiency. The proposed mechanism is evaluated in comprehensive field evaluations in public Long Term Evolution (LTE) networks, where it is able to increase the mean data rate by 194% while simultaneously reducing the average power consumption by up to 54%.

Class imbalance occurs when data classes are not equally represented. Generally, it occurs when some classes represent rare events, while the other classes represent the counterpart of these events. Rare events, especially those that may have a negative impact, often require informed decision-making in a timely manner. However, class imbalance is known to induce a learning bias towards majority classes which implies a poor detection of minority classes. Thus, we propose a new ensemble method to handle class imbalance explicitly at training time. In contrast to existing ensemble methods for class imbalance that use either data driven or randomized approaches for their constructions, our method exploits both directions. On the one hand, ensemble members are built from randomized subsets of training data. On the other hand, we construct different scenarios of class imbalance for the unknown test data. An ensemble is built for each resulting scenario by combining random sampling with the estimation of the relative importance of specific loss functions. Final predictions are generated by a weighted average of each ensemble prediction. As opposed to existing methods, our approach does not try to fix imbalanced data sets. Instead, we show how imbalanced data sets can make classification easier, due to a limited range of true class frequencies. Our procedure promotes diversity among the ensemble members and is not sensitive to specific parameter settings. An experimental demonstration shows, that our new method outperforms or is on par with state-of-the-art ensembles and class imbalance techniques.

Obtaining the distribution of a physical quantity is a frequent objective in experimental physics. In cases where the distribution of the relevant quantity cannot be accessed experimentally, it has to be reconstructed from distributions of correlated quantities that are measured, instead. This reconstruction is called deconvolution. Cherenkov astronomy is a deconvolution use case which studies the energy distribution of cosmic gamma radiation to reason about the characteristics of celestial objects emitting such radiation. We present a novel unified view on deconvolution methods, rephrasing them in the language of data science. Based on our unified formulation, we propose a novel stopping condition that guarantees fast convergence. We compare existing and new methods on synthetic and real-world data, showing that our method converges faster and more accurately than the existing machine learning based approach.

This paper explores the probability of deadline misses for a set of constrained-deadline sporadic soft real-time tasks on uniprocessor platforms. We explore two directions to evaluate the probability whether a job of the task under analysis can finish its execution at (or before) a testing time point t. One approach is based on analytical upper bounds that can be efficiently computed in polynomial time at the price of precision loss for each testing point, derived from the well-known Hoeffding's inequality and the well-known Bernstein's inequality. Another approach convolutes the probability efficiently over multinomial distributions, exploiting a series of state space reduction techniques, i.e., pruning without any loss of precision, and approximations via unifying equivalent classes with a bounded loss of precision. We demonstrate the effectiveness of our approaches in a series of evaluations. Distinct from the convolution-based methods in the literature, which suffer from the high computation demand and are applicable only to task sets with a few tasks, our approaches can scale reasonably without losing much precision in terms of the derived probability of deadline misses.

Boolean matrix factorization (BMF) is a popular and powerful technique for inferring knowledge from data. The mining result is the Boolean product of two matrices, approximating the input dataset. The Boolean product is a disjunction of rank-1 binary matrices, each describing a feature-relation, called pattern, for a group of samples. Yet, there are no guarantees that any of the returned patterns do not actually arise from noise, i.e., are false discoveries. In this paper, we propose and discuss the usage of the false discovery rate in the unsupervised BMF setting. We prove two bounds on the probability that a found pattern is constituted of random Bernoulli-distributed noise. Each bound exploits a specific property of the factorization which minimizes the approximation error—yielding new insights on the minimizers of Boolean matrix factorization. This leads to improved BMF algorithms by replacing heuristic rank selection techniques with a theoretically well-based approach. Our empirical demonstration shows that both bounds deliver excellent results in various practical settings.

Energy-aware system design is an important optimization task for static and mobile Internet of Things (IoT)-based sensor nodes, especially for highly resource-constrained vehicles such as mobile robotic systems. For 4G/5G-based cellular communication systems, the effective transmission power of uplink data transmissions is of crucial importance for the overall system power consumption. Unfortunately, this information is usually hidden within off-the-shelf modems and mobile handsets and can therefore not be exploited for enabling green communication. Moreover, the dynamic transmission power control behavior of the mobile device is not even explicitly modeled in most of the established simulation frameworks. In this paper, we present a novel machine learning-based approach for forecasting the resulting uplink transmission power used for data transmissions based on the available passive network quality indicators and application-level information. The model is derived from comprehensive field measurements of drive tests performed in a public cellular network and can be parameterized for integrating all measurements a given target platform is able to provide into the prediction process. In a comparison of three different machine learning methods, Random-Forest models thoroughly performed best with a mean average error of 3.166 dB. As the absolute sum of errors converges towards zero and falls below 1 dB after 28 predictions in average, the approach is well-suited for long-term power estimations.

The partition function is fundamental for probabilistic graphical models—it is required for inference, parameter estimation, and model selection. Evaluating this function corresponds to discrete integration, namely a weighted sum over an exponentially large set. This task quickly becomes intractable as the dimensionality of the problem increases. We propose an approximation scheme that, for any discrete graphical model whose parameter vector has bounded norm, estimates the partition function with arbitrarily small error. Our algorithm relies on a near minimax optimal polynomial approximation to the potential function and a Clenshaw-Curtis style quadrature. Furthermore, we show that this algorithm can be randomized to split the computation into a high-complexity part and a low-complexity part, where the latter may be carried out on small computational devices. Experiments confirm that the new randomized algorithm is highly accurate if the parameter norm is small, and is otherwise comparable to methods with unbounded error.

Urban traffic gathers increasing interest as cities become bigger, crowded and “smart”. We present a system for het-erogeneous stream processing and crowdsourcing supporting intelligent urban traffic management. Complex events related to traffic congestion (trends) are detected from heterogeneous sources involving fixed sensors mounted on intersections and mobile sensors mounted on public transport vehicles. To deal with data veracity, a crowdsourcing component handles and resolves sensor disagreement. Furthermore, to deal with data sparsity, a traffic modelling component offers information in areas with low sensor coverage. We demonstrate the system with a real-world use-case from Dublin city, Ireland.

Machine learning on resource constrained ubiquitous devices suffers from high energy consumption and slow execution time. In this paper, it is investigated how to modify machine learning algorithms in order to reduce the number of consumed clock cyclesnot by reducing the asymptotic complexity, but by assuming a weaker execution platform. In particular, an integer approximation to the class of undirected graphical models is proposed. Algorithms for inference, maximum-a-posteriori prediction and parameter estimation are presented and approximation error is discussed. In numerical evaluations on synthetic data, the response of the model to several influential properties of the data is investigated. The results on the synthetic data are confirmed with a natural language processing task on an open data set. In addition, the runtime on low-end hardware is regarded. The overall speedup of the new algorithms is at least 2× while overall loss in accuracy is rather small. This allows running probabilistic methods on very small devices, even if they do not contain a processor that is capable of executing floating point arithmetic at all.

The creation of interesting opponents for human players in computer games is an interesting and challenging task. In contrast to up-to-date computer games, e.g. real time strategy games, learning of non-player-character strategies for older games seems to be easier and not that time-consuming. This way, older games, like the famous arcade game Pac-Man, serve as a test bed for the creation of strategies that are fun to play against. The paper at hand uses computational intelligence methods to accomplish this challenge, namely evolutionary algorithms (EA) and artificial neural networks (ANN). The latter are trained on a model of the game whereas the EA learn good behavior by playing. The performance of these two approaches is compared on the original Pac-Man level as well as on other maps with different properties to test the ability of generalizing the learned strategies.