arXiv Computer Science @arxiv_cs@qoto.org

In this short note we derive a relationship between the Bregman divergence from the current policy to the optimal policy and the suboptimality of the current value function in a regularized Markov decision process. This result has implications for multi-task reinforcement learning, offline reinforcement learning, and regret analysis under function approximation, among others.

Recent research has explored using neural networks to reconstruct undersampled magnetic resonance imaging (MRI) data. Because of the complexity of the artifacts in the reconstructed images, there is a need to develop task-based approaches of image quality. Common metrics for evaluating image quality like the normalized root mean squared error (NRMSE) and structural similarity (SSIM) are global metrics which average out impact of subtle features in the images. Using measures of image quality which incorporate a subtle signal for a specific task allow for image quality assessment which locally evaluates the effect of undersampling on a signal. We used a U-Net to reconstruct under-sampled images with 2x, 3x, 4x and 5x fold 1-D undersampling rates. Cross validation was performed for a 500 and a 4000 image training set with both structural similarity (SSIM) and mean squared error (MSE) losses. A two alternative forced choice (2-AFC) observer study was carried out for detecting a subtle signal (small blurred disk) from images with the 4000 image training set. We found that for both loss functions and training set sizes, the human observer performance on the 2-AFC studies led to a choice of a 2x undersampling but the SSIM and NRMSE led to a choice of a 3x undersampling. For this task, SSIM and NRMSE led to an overestimate of the achievable undersampling using a U-Net before a steep loss of image quality when compared to the performance of human observers in the detection of a subtle lesion.

Variational quantum algorithms have been introduced as a promising class of quantum-classical hybrid algorithms that can already be used with the noisy quantum computing hardware available today by employing parameterized quantum circuits. Considering the non-trivial nature of quantum circuit compilation and the subtleties of quantum computing, it is essential to verify that these parameterized circuits have been compiled correctly. Established equivalence checking procedures that handle parameter-free circuits already exist. However, no methodology capable of handling circuits with parameters has been proposed yet. This work fills this gap by showing that verifying the equivalence of parameterized circuits can be achieved in a purely symbolic fashion using an equivalence checking approach based on the ZX-calculus. At the same time, proofs of inequality can be efficiently obtained with conventional methods by taking advantage of the degrees of freedom inherent to parameterized circuits. We implemented the corresponding methods and proved that the resulting methodology is complete. Experimental evaluations (using the entire parametric ansatz circuit library provided by Qiskit as benchmarks) demonstrate the efficacy of the proposed approach. The implementation is open source and publicly available as part of the equivalence checking tool QCEC (https://github.com/cda-tum/qcec) which is part of the Munich Quantum Toolkit (MQT).

Most existing proposals about anaphoric zero pronoun (AZP) resolution regard full mention coreference and AZP resolution as two independent tasks, even though the two tasks are clearly related. The main issues that need tackling to develop a joint model for zero and non-zero mentions are the difference between the two types of arguments (zero pronouns, being null, provide no nominal information) and the lack of annotated datasets of a suitable size in which both types of arguments are annotated for languages other than Chinese and Japanese. In this paper, we introduce two architectures for jointly resolving AZPs and non-AZPs, and evaluate them on Arabic, a language for which, as far as we know, there has been no prior work on joint resolution. Doing this also required creating a new version of the Arabic subset of the standard coreference resolution dataset used for the CoNLL-2012 shared task (Pradhan et al.,2012) in which both zeros and non-zeros are included in a single dataset.

A common paradigm for identifying semantic differences across social and temporal contexts is the use of static word embeddings and their distances. In particular, past work has compared embeddings against "semantic axes" that represent two opposing concepts. We extend this paradigm to BERT embeddings, and construct contextualized axes that mitigate the pitfall where antonyms have neighboring representations. We validate and demonstrate these axes on two people-centric datasets: occupations from Wikipedia, and multi-platform discussions in extremist, men's communities over fourteen years. In both studies, contextualized semantic axes can characterize differences among instances of the same word type. In the latter study, we show that references to women and the contexts around them have become more detestable over time.

The massive increase of data traffic, the widespread proliferation of wireless applications and the full-scale deployment of 5G and the IoT, imply a steep increase in cellular networks energy use, resulting in a significant carbon footprint. This paper presents a comprehensive model to show the interaction between the networking and energy features of the problem and study the economical and technical viability of green networking. Solar equipment, cell zooming, energy management and dynamic user allocation are considered in the upgrading network planning process. We propose a mixed-integer optimization model to minimize long-term capital costs and operational energy expenditures in a heterogeneous on-grid cellular network with different types of base station, including solar. Based on eight scenarios where realistic costs of solar panels, batteries, and inverters were considered, we first found that solar base stations are currently not economically interesting for cellular operators. We next studied the impact of a significant and progressive carbon tax on reducing greenhouse gas emissions (GHG). We found that, at current energy and equipment prices, a carbon tax ten-fold the current value is the only element that could make green base stations economically viable.

This paper investigates the applicability of a recently-proposed nonlinear sparse Bayesian learning (NSBL) algorithm to identify and estimate the complex aerodynamics of limit cycle oscillations. NSBL provides a semi-analytical framework for determining the data-optimal sparse model nested within a (potentially) over-parameterized model. This is particularly relevant to nonlinear dynamical systems where modelling approaches involve the use of physics-based and data-driven components. In such cases, the data-driven components, where analytical descriptions of the physical processes are not readily available, are often prone to overfitting, meaning that the empirical aspects of these models will often involve the calibration of an unnecessarily large number of parameters. While it may be possible to fit the data well, this can become an issue when using these models for predictions in regimes that are different from those where the data was recorded. In view of this, it is desirable to not only calibrate the model parameters, but also to identify the optimal compromise between data-fit and model complexity. In this paper, this is achieved for an aeroelastic system where the structural dynamics are well-known and described by a differential equation model, coupled with a semi-empirical aerodynamic model for laminar separation flutter resulting in low-amplitude limit cycle oscillations. For the purpose of illustrating the benefit of the algorithm, in this paper, we use synthetic data to demonstrate the ability of the algorithm to correctly identify the optimal model and model parameters, given a known data-generating model. The synthetic data are generated from a forward simulation of a known differential equation model with parameters selected so as to mimic the dynamics observed in wind-tunnel experiments.

Much work has been done in multiscale topology optimization for maximum stiffness or minimum compliance design. Such approaches date back to the original homogenization-based work by Bendsøe and Kikuchi from 1988, which lately has been revived due to advances in manufacturing methods like additive manufacturing. Orthotropic microstructures locally oriented in principal stress directions provide for highly efficient stiffness optimal designs, whereas for the pure stiffness objective, porous isotropic microstructures are sub-optimal and hence not useful. It has, however, been postulated and exemplified that isotropic microstructures (infill) may enhance structural buckling stability but this has yet to be directly proven and optimized. In this work, we optimize buckling stability of multiscale structures with isotropic porous infill. To do this, we establish local density dependent Willam-Warnke yield surfaces based on buckling estimates from Bloch-Floquet-based cell analysis to predict local instability of the homogenized materials. These local buckling-based stress constraints are combined with a global buckling criterion to obtain topology optimized designs that take both local and global buckling stability into account. De-homogenized structures with small and large cell sizes confirm validity of the approach and demonstrate huge structural gains as well as time savings compared to standard singlescale approaches.

Objective: A major challenge in closed-loop brain-computer interfaces (BCIs) is finding optimal stimulation patterns as a function of ongoing neural activity for different subjects and objectives. Traditional approaches, such as those currently used for deep brain stimulation, have largely followed a trial- and-error strategy to search for effective open-loop stimulation parameters, a strategy that is inefficient and does not generalize to closed-loop activity-dependent stimulation. Approach: To achieve goal-directed closed-loop neurostimulation, we propose the use of brain co-processors, devices which exploit artificial intelligence (AI) to shape neural activity and bridge injured neural circuits for targeted repair and rehabilitation. Here we investigate a specific type of co-processor called a "neural co-processor" which uses artificial neural networks (ANNs) to learn optimal closed-loop stimulation policies. The co-processor adapts the stimulation policy as the biological circuit itself adapts to the stimulation, achieving a form of brain-device co-adaptation. We tested the neural co-processor's ability to restore function after stroke by simulating a variety of lesions in a previously published cortical model of grasping. Main results: Our results show that a neural co-processor can restore reaching and grasping function after a simulated stroke in a cortical model, achieving recovery towards healthy function in the range 75-90%. Significance: This is the first proof-of-concept demonstration, using computer simulations, of a neural co-processor for activity-dependent closed-loop neurosimulation for optimizing a rehabilitation goal after injury. Our results provide insights on how such co-processors may eventually be developed for in vivo use to learn complex adaptive stimulation policies for a variety of neural rehabilitation and neuroprosthetic applications.

While consolidation strategies form the backbone of many supply chain optimisation problems, exploitation of multi-tier material relationships through consolidation remains an understudied area, despite being a prominent feature of industries that produce complex made-to-order products. In this paper, we propose an optimisation framework for exploiting multi-to-multi relationship between tiers of a supply chain. The resulting formulation is flexible such that quantity discounts, inventory holding and transport costs can be included. The framework introduces a new trade-off between the tiers, resulting in cost reductions at one tier at the expense of increased costs in the other tier, which helps to reduce the overall procurement cost in the supply chain. A mixed integer linear programming model is developed and tested with a range of small to large-scale test problems from aerospace manufacturing. Our comparison to benchmark results show that there is indeed a cost trade-off between two tiers, and that its reduction can be achieved using a holistic approach to reconfiguration. Costs are decreased when second tier fixed ordering costs and the number of machining options increase. Consolidation results in less inventory holding costs for all cases. A number of secondary effects such as simplified supplier selection may also be observed.

This review covers the new developments in machine learning (ML) that are impacting the multi-disciplinary area of aerospace engineering, including fundamental fluid dynamics (experimental and numerical), aerodynamics, acoustics, combustion and structural health monitoring. We review the state of the art, gathering the advantages and challenges of ML methods across different aerospace disciplines and provide our view on future opportunities. The basic concepts and the most relevant strategies for ML are presented together with the most relevant applications in aerospace engineering, revealing that ML is improving aircraft performance and that these techniques will have a large impact in the near future.

Detecting probable Drug Target Interaction (DTI) is a critical task in drug discovery. Conventional DTI studies are expensive, labor-intensive, and take a lot of time, hence there are significant reasons to construct useful computational techniques that may successfully anticipate possible DTIs. Although certain methods have been developed for this cause, numerous interactions are yet to be discovered, and prediction accuracy is still low. To meet these challenges, we propose a DTI prediction model built on molecular structure of drugs and sequence of target proteins. In the proposed model, we use Simplified Molecular Input Line Entry System (SMILES) to create CDK descriptors, Molecular ACCess System (MACCS) fingerprints, Electrotopological state (Estate) fingerprints and amino acid sequences of targets to get Pseudo Amino Acid Composition (PseAAC). We target to evaluate performance of DTI prediction models using CDK descriptors. For comparison, we use benchmark data and evaluate models performance on two widely used fingerprints, MACCS fingerprints and Estate fingerprints. The evaluation of performances shows that CDK descriptors are superior at predicting DTIs. The proposed method also outperforms other previously published techniques significantly.

In collider-based particle and nuclear physics experiments, data are produced at such extreme rates that only a subset can be recorded for later analysis. Typically, algorithms select individual collision events for preservation and store the complete experimental response. A relatively new alternative strategy is to additionally save a partial record for a larger subset of events, allowing for later specific analysis of a larger fraction of events. We propose a strategy that bridges these paradigms by compressing entire events for generic offline analysis but at a lower fidelity. An optimal-transport-based $β$ Variational Autoencoder (VAE) is used to automate the compression and the hyperparameter $β$ controls the compression fidelity. We introduce a new approach for multi-objective learning functions by simultaneously learning a VAE appropriate for all values of $β$ through parameterization. We present an example use case, a di-muon resonance search at the Large Hadron Collider (LHC), where we show that simulated data compressed by our $β$-VAE has enough fidelity to distinguish distinct signal morphologies.

Conventional wisdom in designing the optical switching nodes is rooted in the intuition that when an optical channel crossing an intermediate node, it should be maximally isolated from other optical channels to avoid interference. Such long-established paradigm perceiving the interference of optical channels transiting at the same node as an adversarial factor and should therefore circumvent, albeit reasonable, may leave vast unexplored opportunities. Indeed, rapid advances in all-optical signal processing technologies has brought opportunities to re-define the optical node architecture by upgrading its naive functionalities from simply add/drop and cross-connecting to proactively mixing optical channels in photonic domain. Specifically, all-optical channel (de-) aggregation technologies have been remarkably advancing in recent years, permitting two or more optical channels at lower bit-rate and/or modulation formats could be all-optically aggregated to a single channel of higher-rate and/or higher-order modulation format and vice versa. Such evolutionary technique is poised to disrupt the existing ecosystem for optical network design and planning, and thus necessitates for a radical change to unlock new potentials. In addressing this disruptive idea, we present a new paradigm for future optical networks, namely, optical-processing-enabled networks powered by in-network all-optical mixing capability. We introduce the operational principle of optical channel (de-) aggregation and show how spectrally beneficial such innovative operations could yield by an illustrative example. Next, in order to maximize the aggregation opportunity, we present a mathematical model for optimal routing based on integer linear programming model. Numerical results on the realistic network topology COST239 are provided to quantify the spectral gain of aggregation-aware routing compared to the conventional one.

Traditional (fickle) adversarial examples involve finding a small perturbation that does not change an input's true label but confuses the classifier into outputting a different prediction. Conversely, obstinate adversarial examples occur when an adversary finds a small perturbation that preserves the classifier's prediction but changes the true label of an input. Adversarial training and certified robust training have shown some effectiveness in improving the robustness of machine learnt models to fickle adversarial examples. We show that standard adversarial training methods focused on reducing vulnerability to fickle adversarial examples may make a model more vulnerable to obstinate adversarial examples, with experiments for both natural language inference and paraphrase identification tasks. To counter this phenomenon, we introduce Balanced Adversarial Training, which incorporates contrastive learning to increase robustness against both fickle and obstinate adversarial examples.

Research question: How can we establish an AI support for reading of chest X-rays in clinical routine and which benefits emerge for the clinicians and radiologists. Can it perform 24/7 support for practicing clinicians? 2. Findings: We installed an AI solution for Chest X-ray in a given structure (MVZ Uhlenbrock & Partner, Germany). We could demonstrate the practicability, performance, and benefits in 10 connected clinical sites. 3. Meaning: A commercially available AI solution for the evaluation of Chest X-ray images is able to help radiologists and clinical colleagues 24/7 in a complex environment. The system performs in a robust manner, supporting radiologists and clinical colleagues in their important decisions, in practises and hospitals regardless of the user and X-ray system type producing the image-data.

Neural networks have revolutionized the field of machine learning with increased predictive capability. In addition to improving the predictions of neural networks, there is a simultaneous demand for reliable uncertainty quantification on estimates made by machine learning methods such as neural networks. Bayesian neural networks (BNNs) are an important type of neural network with built-in capability for quantifying uncertainty. This paper discusses aleatoric and epistemic uncertainty in BNNs and how they can be calculated. With an example dataset of images where the goal is to identify the amplitude of an event in the image, it is shown that epistemic uncertainty tends to be lower in images which are well-represented in the training dataset and tends to be high in images which are not well-represented. An algorithm for out-of-distribution (OoD) detection with BNN epistemic uncertainty is introduced along with various experiments demonstrating factors influencing the OoD detection capability in a BNN. The OoD detection capability with epistemic uncertainty is shown to be comparable to the OoD detection in the discriminator network of a generative adversarial network (GAN) with comparable network architecture.

We revisit binary decision trees from the perspective of partitions of the data. We introduce the notion of partitioning function, and we relate it to the growth function and to the VC dimension. We consider three types of features: real-valued, categorical ordinal and categorical nominal, with different split rules for each. For each feature type, we upper bound the partitioning function of the class of decision stumps before extending the bounds to the class of general decision tree (of any fixed structure) using a recursive approach. Using these new results, we are able to find the exact VC dimension of decision stumps on examples of $\ell$ real-valued features, which is given by the largest integer $d$ such that $2\ell \ge \binom{d}{\lfloor\frac{d}{2}\rfloor}$. Furthermore, we show that the VC dimension of a binary tree structure with $L_T$ leaves on examples of $\ell$ real-valued features is in $O(L_T \log(L_T\ell))$. Finally, we elaborate a pruning algorithm based on these results that performs better than the cost-complexity and reduced-error pruning algorithms on a number of data sets, with the advantage that no cross-validation is required.

Topology optimization (TO) is a method of deriving an optimal design that satisfies a given load and boundary conditions within a design domain. This method enables effective design without initial design, but has been limited in use due to high computational costs. At the same time, machine learning (ML) methodology including deep learning has made great progress in the 21st century, and accordingly, many studies have been conducted to enable effective and rapid optimization by applying ML to TO. Therefore, this study reviews and analyzes previous research on ML-based TO (MLTO). Two different perspectives of MLTO are used to review studies: (1) TO and (2) ML perspectives. The TO perspective addresses "why" to use ML for TO, while the ML perspective addresses "how" to apply ML to TO. In addition, the limitations of current MLTO research and future research directions are examined.

Applications of Structural Health Monitoring (SHM) combined with Machine Learning (ML) techniques enhance real-time performance tracking and increase structural integrity awareness of civil, aerospace and automotive infrastructures. This SHM-ML synergy has gained popularity in the last years thanks to the anticipation of maintenance provided by arising ML algorithms and their ability of handling large quantities of data and considering their influence in the problem. In this paper we develop a novel ML nearest-neighbors-alike algorithm based on the principle of maximum entropy to predict fatigue damage (Palmgren-Miner index) in composite materials by processing the signals of Lamb Waves -- a non-destructive SHM technique -- with other meaningful features such as layup parameters and stiffness matrices calculated from the Classical Laminate Theory (CLT). The full data analysis cycle is applied to a dataset of delamination experiments in composites. The predictions achieve a good level of accuracy, similar to other ML algorithms, e.g. Neural Networks or Gradient-Boosted Trees, and computation times are of the same order of magnitude. The key advantages of our proposal are: (1) The automatic determination of all the parameters involved in the prediction, so no hyperparameters have to be set beforehand, which saves time devoted to hypertuning the model and also represents an advantage for autonomous, self-supervised SHM. (2) No training is required, which, in an \textit{online learning} context where streams of data are fed continuously to the model, avoids repeated training -- essential for reliable real-time, continuous monitoring.