Open Challenges Research Articles

Applied AI in Defect Detection for Additive Manufacturing: Current Literature, Metrics, Datasets, and Open Challenges

Applied AI in Defect Detection for Additive Manufacturing: Current Literature, Metrics, Datasets, and Open Challenges

Deep Learning Techniques for Detecting and Segmenting Text in Natural Scene Images: Review

Text detection and segmentation in natural scene images is an active research problem in computer vision and document analysis. Unlike scanned documents, scene text exhibits significant diversity in appearance, orientation, scale, font, and lighting conditions. In this review, survey the current state-of-the-art in techniques and methodologies aimed at detecting and segmenting text regions from images of natural scenes are presented. Both traditional approaches using hand-crafted features as well as modern data-driven deep learning methods will be discussed. The review will analyze common datasets, evaluation protocols and metrics used for benchmarking. Limitations of existing methods and open challenges in handling multilingual text, curved text, and efficiency will be highlighted. Promising future directions towards robust and generalizable scene text extraction systems will be identified. In summary, the review will provide a comprehensive overview of the advances, remaining challenges and future opportunities in developing automated systems for detecting and segmenting text in unconstrained natural images.

HAPS for 6G Networks: Potential Use Cases, Open Challenges, and Possible Solutions

HAPS for 6G Networks: Potential Use Cases, Open Challenges, and Possible Solutions

Matching workloads ro systems with deep reinforcement learning

Along with the evolution of computer microarchitecture over the years, the number of dies, cores, and embedded multi-die interconnect bridges has grown. Optimizing the workload running on a central processing unit (CPU) to improve the computer performance has become a challenge. Matching workloads to systems with optimal system configurations to achieve the desired system performance is an open challenge in both academic and industrial research. In this paper, we propose two reinforcement learning (RL) approaches, a deep reinforcement learning (DRL) approach and an evolutionary deep reinforcement learning (EDRL) approach, to find an optimal system configuration for a given computer workload with a system performance objective. The experimental results demonstrate that both approaches can determine the optimal system configuration with the desired performance objective. The comparison studies illustrate that the DRL approach outperforms the standard RL approaches. In the future, these DRL approaches can be leveraged in system performance auto-tuning studies.

Open Source Intelligence Opportunities and Challenges: a Review

Open Source Intelligence Opportunities and Challenges: a Review

Optimal sampling of spatial patterns improves deep learning-based early warning signals of critical transitions.

Complex spatio-temporal systems like lakes, forests and climate systems exhibit alternative stable states. In such systems, as the threshold value of the driver is crossed, the system may experience a sudden (discontinuous) transition or smooth (continuous) transition to an undesired steady state. Theories predict that changes in the structure of the underlying spatial patterns precede such transitions. While there has been a large body of research on identifying early warning signals of critical transitions, the problem of forecasting the type of transitions (sudden versus smooth) remains an open challenge. We address this gap by developing an advanced machine learning (ML) toolkit that serves as an early warning indicator of spatio-temporal critical transitions, Spatial Early Warning Signal Network (S-EWSNet). ML models typically resemble a black box and do not allow envisioning what the model learns in discerning the labels. Here, instead of naively relying upon the deep learning model, we let the deep neural network learn the latent features characteristic of transitions via an optimal sampling strategy (OSS) of spatial patterns. The S-EWSNet is trained on data from a stochastic cellular automata model deploying the OSS, providing an early warning indicator of transitions while detecting its type in simulated and empirical samples.

Integrated Sensing and Communications: Recent Advances and Ten Open Challenges

Integrated Sensing and Communications: Recent Advances and Ten Open Challenges

FCM-DCS: Fuzzy C means distorted contour-based segmentation model for breast cancer detection

Breast cancer (BC) is a commonly diagnosed cancer among women nowadays. The cancer cells in the breast tissues are known as BC. Comprehensive research on early-stage BC detection helped to increase the survival rate and reduce the amortality rate associated with this disease. Mammogram scan analysis is a commonly used breast tissue visualization method. This image data is analyzed adequately for accurate BC diagnosis. Region of interest (ROI) identification is crucial in an image-based BC detection system. The ROI detection helps to segment the cancer tissues from the mammogram images by analyzing the heterogeneity among cancerous and normal breast tissues. Early-stage BC issues have homogeneous features as normal breast tissues. So, it is an open challenge for the researchers to develop a more accurate segmentation method during the automatic BC stages detection system. This study introduced fuzzy C means (FCM) distorted contour-based segmentation (FCM DCS) method to address the real detection issues in present studies. It uses the distorted contour (DC) based method to identify the contour of the cancer tissue from mammogram images. The DC method is performed with the help of FCM to identify the cancer tissues. Moreover, a histogram and adaptive equalization method were utilized to reduce image noise and preserve the edge features. The result analysis shows that the FCM-DC methods achieved a maximum accuracy rate (98.76 %) than comparison methods in BC detection.

Dynamic Inverse Design of Broadband Metasurfaces with Synthetical Neural Networks

AbstractFor over 35 years of research, the debate about the systematic compositionality of neural networks remains unchanged, arguing that existing artificial neural networks are inadequate cognitive models. Recent advancements in deep learning have significantly shaped the landscape of popular domains, however, the systematic combination of previously trained neural networks remains an open challenge. This study presents how to dynamically synthesize a neural network for the design of broadband electromagnetic metasurfaces. The underlying mechanism relies on an assembly network to adaptively integrate pre‐trained inherited networks in a transparent manner that corresponds to the metasurface assembly in physical space. This framework is poised to curtail data requirements and augment network flexibility, promising heightened practical utility in complex composition‐based tasks. Importantly, the intricate coupling effects between different metasurface segments are accurately captured. The approach for two broadband metasurface inverse design problems is exemplified, reaching accuracies of 96.7% and 95.5%. Along the way, the importance of suitably formatting the spectral data is highlighted to capture sharp spectral features. This study marks a significant leap forward in inheriting pre‐existing knowledge in neural‐network‐based inverse design, improving its adaptability for applications involving dynamically evolving tasks.

Model predictive-based DNN control model for automated steering deployed on FPGA using an automatic IP generator tool

With the increase in the non-linearity and complexity of the driving system’s environment, developing and optimizing related applications is becoming more crucial and remains an open challenge for researchers and automotive companies alike. Model predictive control (MPC) is a well-known classic control strategy used to solve online optimization problems. MPC is computationally expensive and resource-consuming. Recently, machine learning has become an effective alternative to classical control systems. This paper provides a developed deep neural network (DNN)-based control strategy for automated steering deployed on FPGA. The DNN model was designed and trained based on the behavior of the traditional MPC controller. The performance of the DNN model is evaluated compared to the performance of the designed MPC which already proved its merit in automated driving task. A new automatic intellectual property generator based on the Xilinx system generator (XSG) has been developed, not only to perform the deployment but also to optimize it. The performance was evaluated based on the ability of the controllers to drive the lateral deviation and yaw angle of the vehicle to be as close as possible to zero. The DNN model was implemented on FPGA using two different data types, fixed-point and floating-point, in order to evaluate the efficiency in the terms of performance and resource consumption. The obtained results show that the suggested DNN model provided a satisfactory performance and successfully imitated the behavior of the traditional MPC with a very small root mean square error (RMSE = 0.011228 rad). Additionally, the results show that the deployments using fixed-point data greatly reduced resource consumption compared to the floating-point data type while maintaining satisfactory performance and meeting the safety conditions

Mastering citizen involvement in international open innovation challenges: Interim lessons learned of the FRANCIS project

EU-funded project FRANCIS focuses on involving citizens in the development of so-called Frugal Innovations, i.e. simple, affordable and sustainable solutions. It does so by organising open-innovations challenges that are managed through an IT-platform and supported by various recruitment and ideation formats. Usually, Citizen Science only involves citizens for minor tasks such as data collection. FRANCIS jump ahead in science is that the project invites them to propose and work on new solutions that have a real chance to enter the market. Before starting the challenges, the team of FRANCIS already made some assumptions about how to encourage and support participants in the best possible way based on existing publications and first findings from primary research. Now that the first challenge is running, the team had the opportunity to check whether they have proven to be true so far. This paper presents the interim lessons learned and details the way forward

Towards long-tailed, multi-label disease classification from chest X-ray: Overview of the CXR-LT challenge

Many real-world image recognition problems, such as diagnostic medical imaging exams, are “long-tailed” – there are a few common findings followed by many more relatively rare conditions. In chest radiography, diagnosis is both a long-tailed and multi-label problem, as patients often present with multiple findings simultaneously. While researchers have begun to study the problem of long-tailed learning in medical image recognition, few have studied the interaction of label imbalance and label co-occurrence posed by long-tailed, multi-label disease classification. To engage with the research community on this emerging topic, we conducted an open challenge, CXR-LT, on long-tailed, multi-label thorax disease classification from chest X-rays (CXRs). We publicly release a large-scale benchmark dataset of over 350,000 CXRs, each labeled with at least one of 26 clinical findings following a long-tailed distribution. We synthesize common themes of top-performing solutions, providing practical recommendations for long-tailed, multi-label medical image classification. Finally, we use these insights to propose a path forward involving vision-language foundation models for few- and zero-shot disease classification.

Nanocrystal Assemblies: Current Advances and Open Problems.

We explore the potential of nanocrystals (a term used equivalently to nanoparticles) as building blocks for nanomaterials, and the current advances and open challenges for fundamental science developments and applications. Nanocrystal assemblies are inherently multiscale, and the generation of revolutionary material properties requires a precise understanding of the relationship between structure and function, the former being determined by classical effects and the latter often by quantum effects. With an emphasis on theory and computation, we discuss challenges that hamper current assembly strategies and to what extent nanocrystal assemblies represent thermodynamic equilibrium or kinetically trapped metastable states. We also examine dynamic effects and optimization of assembly protocols. Finally, we discuss promising material functions and examples of their realization with nanocrystal assemblies.

Heterogeneous integration of spin-photon interfaces with a CMOS platform.

Colour centres in diamond have emerged as a leading solid-state platform for advancing quantum technologies, satisfying the DiVincenzo criteria1 and recently achieving quantum advantage in secret key distribution2. Blueprint studies3-5 indicate that general-purpose quantum computing using local quantum communication networks will require millions of physical qubits to encode thousands of logical qubits, presenting an open scalability challenge. Here we introduce a modular quantum system-on-chip (QSoC) architecture that integrates thousands of individually addressable tin-vacancy spin qubits in two-dimensional arrays of quantum microchiplets into an application-specific integrated circuit designed for cryogenic control. We demonstrate crucial fabrication steps and architectural subcomponents, including QSoC transfer by means of a 'lock-and-release' method for large-scale heterogeneous integration, high-throughput spin-qubit calibration and spectral tuning, and efficient spin state preparation and measurement. This QSoC architecture supports full connectivity for quantum memory arrays by spectral tuning across spin-photon frequency channels. Design studies building on these measurements indicate further scaling potential by means of increased qubit density, larger QSoC active regions and optical networking across QSoC modules.

Anomaly Detection in Dynamic Graphs: A Comprehensive Survey

This survey paper presents a comprehensive and conceptual overview of anomaly detection using dynamic graphs. We focus on existing graph-based anomaly detection (AD) techniques and their applications to dynamic networks. The contributions of this survey paper include the following: i) a comparative study of existing surveys on anomaly detection; ii) a D ynamic G raph-based A nomaly D etection ( DGAD ) review framework in which approaches for detecting anomalies in dynamic graphs are grouped based on traditional machine-learning models, matrix transformations, probabilistic approaches, and deep-learning approaches; iii) a discussion of graphically representing both discrete and dynamic networks; and iv) a discussion of the advantages of graph-based techniques for capturing the relational structure and complex interactions in dynamic graph data. Finally, this work identifies the potential challenges and future directions for detecting anomalies in dynamic networks. This DGAD survey approach aims to provide a valuable resource for researchers and practitioners by summarizing the strengths and limitations of each approach, highlighting current research trends, and identifying open challenges. In doing so, it can guide future research efforts and promote advancements in anomaly detection in dynamic graphs.

Frequency-Histogram Coarse Graining in Elementary and 2-Dimensional Cellular Automata

Cellular automata and other discrete dynamical systems have long been studied as models of emergent complexity. Recently, neural cellular automata have been proposed as models to investigate the emerge of a more general artificial intelligence, thanks to their propensity to support properties such as self-organization, emergence, and open-endedness. However, understanding emergent complexity in large scale systems is an open challenge. How can the important computations leading to emergent complex structures and behaviors be identified? In this work, we systematically investigate a form of dimensionality reduction for 1-dimensional and 2-dimensional cellular automata based on coarse-graining of macrostates into smaller blocks. We discuss selected examples and provide the entire exploration of coarse graining with different filtering levels in the appendix (available also digitally at this link: https://s4nyam.github.io/eca88/. We argue that being able to capture emergent complexity in AI systems may pave the way to open-ended evolution, a plausible path to reach artificial general intelligence.

Empirical uncertainty evaluation for the pose of a kinematic LiDAR-based multi-sensor system

Abstract Kinematic multi-sensor systems (MSS) describe their movements through six-degree-of-freedom trajectories, which are often evaluated primarily for accuracy. However, understanding their self-reported uncertainty is crucial, especially when operating in diverse environments like urban, industrial, or natural settings. This is important, so the following algorithms can provide correct and safe decisions, i.e. for autonomous driving. In the context of localization, light detection and ranging sensors (LiDARs) are widely applied for tasks such as generating, updating, and integrating information from maps supporting other sensors to estimate trajectories. However, popular low-cost LiDARs deviate from other geodetic sensors in their uncertainty modeling. This paper therefore demonstrates the uncertainty evaluation of a LiDAR-based MSS localizing itself using an inertial measurement unit (IMU) and matching LiDAR observations to a known map. The necessary steps for accomplishing the sensor data fusion in a novel Error State Kalman filter (ESKF) will be presented considering the influences of the sensor uncertainties and their combination. The results provide new insights into the impact of random and systematic deviations resulting from parameters and their uncertainties established in prior calibrations. The evaluation is done using the Mahalanobis distance to consider the deviations of the trajectory from the ground truth weighted by the self-reported uncertainty, and to evaluate the consistency in hypothesis testing. The evaluation is performed using a real data set obtained from an MSS consisting of a tactical grade IMU and a Velodyne Puck in combination with reference data by a Laser Tracker in a laboratory environment. The data set consists of measurements for calibrations and multiple kinematic experiments. In the first step, the data set is simulated based on the Laser Tracker measurements to provide a baseline for the results under assumed perfect corrections. In comparison, the results using a more realistic simulated data set and the real IMU and LiDAR measurements provide deviations about a factor of five higher leading to an inconsistent estimation. The results offer insights into the open challenges related to the assumptions for integrating low-cost LiDARs in MSSs.

In vivo demonstration of globotriaosylceramide brain accumulation in Fabry Disease using MR Relaxometry

PurposeHow to measure brain globotriaosylceramide (Gb3) accumulation in Fabry Disease (FD) patients in-vivo is still an open challenge. The objective of this study is to provide a quantitative, non-invasive demonstration of this phenomenon using quantitative MRI (qMRI).MethodsIn this retrospective, monocentric cross-sectional study conducted from November 2015 to July 2018, FD patients and healthy controls (HC) underwent an MRI scan with a relaxometry protocol to compute longitudinal relaxation rate (R1) maps to evaluate gray (GM) and white matter (WM) lipid accumulation. In a subgroup of 22 FD patients, clinical (FAbry STabilization indEX -FASTEX- score) and biochemical (residual α-galactosidase activity) variables were correlated with MRI data. Quantitative maps were analyzed at both global (“bulk” analysis) and regional (“voxel-wise” analysis) levels.ResultsData were obtained from 42 FD patients (mean age = 42.4 ± 12.9, M/F = 16/26) and 49 HC (mean age = 42.3 ± 16.3, M/F = 28/21). Compared to HC, FD patients showed a widespread increase in R1 values encompassing both GM (pFWE = 0.02) and WM (pFWE = 0.02) structures. While no correlations were found between increased R1 values and FASTEX score, a significant negative correlation emerged between residual enzymatic activity levels and R1 values in GM (r = -0.57, p = 0.008) and WM (r = -0.49, p = 0.03).ConclusionsWe demonstrated the feasibility and clinical relevance of non-invasively assessing cerebral Gb3 accumulation in FD using MRI. R1 mapping might be used as an in-vivo quantitative neuroimaging biomarker in FD patients.

Causality for Trustworthy Artificial Intelligence: Status, Challenges and Perspectives

Causal inference is the idea of cause-and-effect; this fundamental area of sciences can be applied to problem space associated with Newton’s laws or the devastating COVID-19 pandemic. The cause explains the “why” whereas the effect describes the “what”. The domain itself encompasses a plethora of disciplines from statistics and computer science to economics and philosophy. Recent advancements in machine learning (ML) and artificial intelligence (AI) systems, have nourished a renewed interest in identifying and estimating the cause-and-effect relationship from the substantial amount of available observational data. This has resulted in various new studies aimed at providing novel methods for identifying and estimating causal inference. We include a detailed taxonomy of causal inference frameworks, methods, and evaluation. An overview of causality for security is also provided. Open challenges are detailed, and approaches for evaluating the robustness of causal inference methods are described. This paper aims to provide a comprehensive survey on such studies of causality. We provide an in-depth review of causality frameworks, and describe the different methods.

Distributed training of CosPlace for large-scale visual place recognition.

Visual place recognition (VPR) is a popular computer vision task aimed at recognizing the geographic location of a visual query, usually within a tolerance of a few meters. Modern approaches address VPR from an image retrieval standpoint using a kNN on top of embeddings extracted by a deep neural network from both the query and images in a database. Although most of these approaches rely on contrastive learning, which limits their ability to be trained on large-scale datasets (due to mining), the recently reported CosPlace proposes an alternative training paradigm using a classification task as the proxy. This has been shown to be effective in expanding the potential of VPR models to learn from large-scale and fine-grained datasets. In this work, we experimentally analyze CosPlace from a continual learning perspective and show that its sequential training procedure leads to suboptimal results. As a solution, we propose a different formulation that not only solves the pitfalls of the original training strategy effectively but also enables faster and more efficient distributed training. Finally, we discuss the open challenges in further speeding up large-scale image retrieval for VPR.