High-definition (HD) maps provide precise environmental information essential for prediction and planning in autonomous driving (AD) systems. Due to the high cost of labeling and maintenance, recent research has turned to online HD map construction using onboard sensor data, offering wider coverage and more timely updates for autonomous vehicles (AVs). However, the robustness of online map construction under adversarial conditions remains underexplored. In this paper, we present a systematic vulnerability analysis of online map construction models, which reveals that these models exhibit an inherent bias toward predicting symmetric road structures. In asymmetric scenes like forks or merges, this bias often causes the model to mistakenly predict a straight boundary that mirrors the opposite side. We demonstrate that this vulnerability persists in the real-world and can be reliably triggered by obstruction or targeted interference. Leveraging this vulnerability, we propose a novel two-stage attack framework capable of manipulating online constructed maps. First, our method identifies vulnerable asymmetric scenes along the victim AV's potential route. Then, we optimize the location and pattern of camera-blinding attacks and adversarial patch attacks. Evaluations on a public AD dataset demonstrate that our attacks can degrade mapping accuracy by up to 9.9% in average precision, render up to 44% of targeted routes unreachable, and increase unsafe planned trajectory rates—colliding with real-world road boundaries—by up to 27%. These attacks are also validated on a real-world testbed vehicle. We further analyze root causes of the symmetry bias, attributing them to training data imbalance, model architecture, and map element representation. Based on these findings, we propose asymmetric data fine-tuning as a targeted defense, which significantly improves model robustness. To the best of our knowledge, this study presents the first vulnerability assessment of online map construction models and introduces the first digital and physical attack against them.

This work introduces a novel Transposed Stochastic Computing-in-Memory (TSCIM) macro designed to enhance the efficiency of on-chip training and inference. The macro incorporates a novel stochastic quantization strategy and utilizes a transposed separated wordline SRAM to enable multi-bit signed MAC operations. Furthermore, a stochastic adder tree is utilized to minimize area and power consumption overhead. The design includes a 4Kb SRAM CIM macro implemented in 28 nm CMOS technology. Simulation results show that the power consumption of the stochastic accumulation circuit (SAC) is reduced by 63.6%, while the area overhead is decreased by a factor of 7.73 compared to designs using full adder (FA) adder trees. Additionally, the computation latency is decreased by 16× compared to traditional stochastic circuits. The TSCIM macro can achieve a peak energy efficiency of 63.02 TOPS/W and an area efficiency of 15.54 TOPS/mm2.

Adversarial attacks with real objects or lasers on car-borne LiDAR-based object detection are concerning. The existing defense approaches are often designed to address specific attacks and short of considering adaptive attackers who may adapt based on all available information about the deployed defense to maximize attack effect. This paper proposes Hyper3Def, a new defense for the function of detecting vehicle objects, which uses a Hypernet to generate an ensemble of multiple new detection models when needed at run time. The detection results of these models are fused to give the final result. As a dynamic defense, Hyper3Def revokes an important basis of the adaptive attack, i.e., the object detection model is needed to plan effective adversarial perturbations. Evaluation based on open data and real-world experiments with embedded system implementation show that, when confronting adaptive attacks, Hyper3Def outperforms various baseline defenses including the adversarial training, which is often cited as the state of the art.

Computing in memory (CIM), which alleviates the need to transfer a large amount of data between processor and memory, significantly reducing latency and energy consumption, is a promising new computing architecture for addressing the von Neumann bottleneck problem. This article proposes a CIM array structure composed of self-recycling 10T static random access memory (SRAM) cells, which can realize orthogonal data writing, and multiple Boolean logical operations for the entire array. The self-recycling and full-array activation characteristics are extremely suitable for accelerating diverse data processing algorithms such as the Advanced Encryption Standard (AES). A 4-kb SRAM is implemented in 55-nm CMOS technology to verify the effectiveness of the design. Compared with other state-of-the-art architectures, the throughput and the operating frequency of the proposed CIM macro are increased to 843 GOPS/kb ($2.64\times $) and 823.7 MHz ($2.6\times $), respectively. The energy efficiency reaches 246.9 TOPS/W. When applied to the AES, the energy consumption is 35.77% less than the digital CIM architecture that is not self-recycling.

To create a future society based on cyber-physical systems, we need real-time digital twins made with cameras and sensors. The challenge is making this energy-efficient. A model in [1] suggests dividing video analysis tasks among terminals, edge servers, and cloud servers to minimize power consumption. This paper addresses energy optimization in dynamic environments with a two-level approach: detecting and categorizing environmental changes (LEC and SEC) and using a two-level adaptive evolutionary algorithm (TAEA) to make corresponding adjustments. A case study with a differential evolution algorithm demonstrates its effectiveness in minimizing energy costs and improving processing accuracy and latency violations.

Network robustness, which refers to a network’s ability to withstand malicious attacks on its vertices and edges, is critical across various natural and industrial domains. This paper delves into the assessment of network robustness through machine learning-based approaches, with a specific focus on structure-based representations and graph embeddings. The evaluation encompasses both synthetic and real-world networks, and three types of representation paradigms are scrutinized: 1) structure-based representations, including adjacency, incidence, and modularity matrices, 2) graph embeddings, including learning feature representation (LFR), DeepWalk, large-scale information network embedding (LINE), node2vec, structural deep network embedding (SDNE), and struc2vec, and 3) one-dimensional graph embeddings. The findings underscore the preference of convolutional neural networks (CNNs) with structure-based representations, highlighting the efficacy of adjacency and modularity matrices. While graph embeddings showcase versatility, their overall performance is comparatively lower, emphasizing the crucial role of representation complexity. This study contributes valuable insights into robustness evaluation methodologies and underscores the significance of tailored graph representations.

Despite various measures across different engineering and social systems, network robustness remains crucial for resisting random faults and malicious attacks. In this study, robustness refers to the ability of a network to maintain its functionality after a part of the network has failed. Existing methods assess network robustness using attack simulations, spectral measures, or deep neural networks (DNNs), which return a single metric as a result. Evaluating network robustness is technically challenging, while evaluating a single metric is practically insufficient. This article proposes a multitask analysis system based on the graph isomorphism network (GIN) model, abbreviated as GIN-MAS. First, a destruction-based robustness metric is formulated using the destruction threshold of the examined network. A multitask learning approach is taken to learn the network robustness metrics, including connectivity robustness, controllability robustness, destruction threshold, and the maximum number of connected components. Then, a five-layer GIN is constructed for evaluating the aforementioned four robustness metrics simultaneously. Finally, extensive experimental studies reveal that 1) GIN-MAS outperforms nine other methods, including three state-of-the-art convolutional neural network (CNN)-based robustness evaluators, with lower prediction errors for both known and unknown datasets from various directed and undirected, synthetic, and real-world networks; 2) the multitask learning scheme is not only capable of handling multiple tasks simultaneously but more importantly it enables the parameter and knowledge sharing across tasks, thus preventing overfitting and enhancing the performances; and 3) GIN-MAS performs multitasks significantly faster than other single-task evaluators. The excellent performance of GIN-MAS suggests that more powerful DNNs have great potentials for analyzing more complicated and comprehensive robustness evaluation tasks.

Voltage stability assessment is essential for maintaining reliable power grid operations. Stability assessment approaches using deep learning address the shortfalls of the traditional time-domain simulation-based approaches caused by increased system complexity. However, deep learning models are shown to be vulnerable to adversarial examples in the field of computer vision. While this vulnerability has been noticed by the power grid cybersecurity research, the domain-specific analysis on the requirements imposed upon effective attack implementation is still lacking. Although these attack requirements are usually reasonable in computer vision tasks, they can be stringent in the context of power grids. In this paper, we conduct a systematic investigation on the attack requirements and credibility of six representative adversarial example attacks based on a voltage stability assessment application for the New England 10-machine 39-bus power system. We show that (1) compromising about half the transmission system buses’ voltage traces is a rule-of-thumb attack requirement; (2) the universal adversarial perturbations regardless of the original clean voltage trajectory possess the same credibility as the widely studied false data injection attacks on power grid state estimation, while the input-specific adversarial perturbations are less credible; (3) the prevailing strong adversarial training thwarts the universal perturbations but fails in defending certain input-specific perturbations. To advance defense to cope with both universal and input-specific adversarial examples, we propose a new approach that simultaneously estimates the predictive uncertainty of any given input of voltage trajectory and thwarts the attacks effectively.

Network robustness refers to the ability of a network to continue its functioning against malicious attacks, which is critical for various natural and industrial networks. Network robustness can be quantitatively measured by a sequence of values that record the remaining functionality after a sequential node- or edge-removal attacks. Robustness evaluations are traditionally determined by attack simulations, which are computationally very time-consuming and sometimes practically infeasible. The convolutional neural network (CNN)-based prediction provides a cost-efficient approach to fast evaluating the network robustness. In this article, the prediction performances of the learning feature representation-based CNN (LFR-CNN) and PATCHY-SAN methods are compared through extensively empirical experiments. Specifically, three distributions of network size in the training data are investigated, including the uniform, Gaussian, and extra distributions. The relationship between the CNN input size and the dimension of the evaluated network is studied. Extensive experimental results reveal that compared to the training data of uniform distribution, the Gaussian and extra distributions can significantly improve both the prediction performance and the generalizability, for both LFR-CNN and PATCHY-SAN, and for various functionality robustness. The extension ability of LFR-CNN is significantly better than PATCHY-SAN, verified by extensive comparisons on predicting the robustness of unseen networks. In general, LFR-CNN outperforms PATCHY-SAN, and thus LFR-CNN is recommended over PATCHY-SAN. However, since both LFR-CNN and PATCHY-SAN have advantages for different scenarios, the optimal settings of the input size of CNN are recommended under different configurations.

Multi-modal fusion has shown initial promising results for object detection of autonomous driving perception. However, many existing fusion schemes do not consider the quality of each fusion input and may suffer from adverse conditions on one or more sensors. While predictive uncertainty has been applied to characterize single-modal object detection performance at run time, incorporating uncertainties into the multi-modal fusion still lacks effective solutions due primarily to the uncertainty’s cross-modal incomparability and distinct sensitivities to various adverse conditions. To fill this gap, this paper proposes Uncertainty-Encoded Mixture-of-Experts (UMoE) that explicitly incorporates single-modal uncertainties into LiDAR-camera fusion. UMoE uses individual expert network to process each sensor’s detection result together with encoded uncertainty. Then, the expert networks’ outputs are analyzed by a gating network to determine the fusion weights. The proposed UMoE module can be integrated into any proposal fusion pipeline. Evaluation shows that UMoE achieves a maximum of 10.67%, 3.17%, and 5.40% performance gain compared with the state-of-the-art proposal-level multi-modal object detectors under extreme weather, adversarial, and blinding attack scenarios.

Many natural and engineering systems can be modeled and represented in the forms of graph data, and then studied using graph theory and network analysis tools. Graph representation learning aims at generating lower-dimensional representations from higher-dimensional graph data, which is a crucial step that facilitates the follow-up tasks, such as node and graph classifications. In this paper, we present a simple but effective graph representation learning method, namely the pyramid pooling-based local profile (PPLP), which enables local nodal profiles to be transformed into a graph representation, with multi-scale features extracted. PPLP can be either embedded into a graph neural network as the readout layer, or perform independently as a graph embedding algorithm. The resultant representations of PPLP are for graph-level tasks. PPLP is experimentally tested by performing graph classification tasks on ten representative datasets, either as the readout layer of different graph neural networks, or as an independent graph embedding algorithm. Experimental results demonstrate that: 1) when embedded into graph neural networks, PPLP outperforms the widely-used global pooling-based readout methods; 2) as an independent graph embedding algorithm, PPLP performs fairly good, especially on the social network datasets. The investigation confirms PPLP as a simple but promising method for graph-level tasks.

This paper addresses the robustness of a network to sustain its connectivity and controllability against malicious attacks. This kind of network robustness is typically measured by the time-consuming attack simulation, which returns a sequence of values that record the remaining connectivity and controllability after a sequence of node- or edge-removal attacks. For improvement, this paper develops an efficient framework for network robustness prediction, the spatial pyramid pooling convolutional neural network (SPP-CNN). The new framework installs a spatial pyramid pooling layer between the convolutional and fully-connected layers, overcoming the common mismatch issue in the CNN-based prediction approaches and extending its generalizability. Extensive experiments are carried out by comparing SPP-CNN with three state-of-the-art robustness predictors, namely one CNN-based and two graph neural networks-based frameworks. Synthetic and real-world networks, both directed and undirected, are investigated. Experimental results demonstrate that the proposed SPP-CNN achieves better prediction performances and better generalizability for both cases of known and unknown datasets, with significantly lower time-consumption, than its counterparts.

Network robustness is critical for various societal and industrial networks against malicious attacks. In particular, connectivity robustness and controllability robustness reflect how well a networked system can maintain its connectedness and controllability against destructive attacks, which can be quantified by a sequence of values that record the remaining connectivity and controllability of the network after a sequence of node- or edge-removal attacks. Traditionally, robustness is determined by attack simulations, which are computationally very time-consuming or even practically infeasible for large-scale networks. In this article, an improved method for network robustness prediction is developed based on learning feature representation using the convolutional neural network (LFR-CNN). In this scheme, the higher-dimensional network data are compressed into lower-dimensional representations, which are then passed to a convolutional neural network to perform robustness prediction. Extensive experimental studies on both synthetic and real-world networks, both directed and undirected, demonstrate that: 1) the proposed LFR-CNN performs better than other two state-of-the-art prediction methods, with significantly smaller prediction errors; 2) LFR-CNN is insensitive to the variation of the input network size, which significantly extends its applicability; 3) although LFR-CNN needs more time to perform feature learning, it can achieve accurate prediction faster than attack simulations; and 4) LFR-CNN not only accurately predicts the network robustness, but also provides a good indicator for connectivity robustness, better than the classical spectral measures.

Connectivity and controllability of a complex network are two important issues that guarantee a networked system to function. Robustness of connectivity and controllability guarantees the system to function properly and stably under various malicious attacks. Evaluating network robustness using attack simulations is time consuming, while the convolutional neural network (CNN)-based prediction approach provides a cost-efficient method to approximate the network robustness. In this paper, we investigate the performance of CNN-based approaches for connectivity and controllability robustness prediction, when partial network information is missing, namely the adjacency matrix is incomplete. Extensive experimental studies are carried out. A threshold is explored that if a total amount of more than 7.29% information is lost, the performance of CNN-based prediction will be significantly degenerated for all cases in the experiments. Two scenarios of missing edge representations are compared, 1) a missing edge is marked ‘no edge’ in the input for prediction, and 2) a missing edge is denoted using a special marker of ‘unknown’. Experimental results reveal that the first representation is misleading to the CNN-based predictors.

In recent years, many deep learning models have been adopted in autonomous driving. At the same time, these models introduce new vulnerabilities that may compromise the safety of autonomous vehicles. Specifically, recent studies have demonstrated that adversarial attacks can cause a significant decline in detection precision of deep learning-based 3-D object detection models. Although driving safety is the ultimate concern for autonomous driving, there is no comprehensive study on the linkage between the performance of deep learning models and the driving safety of autonomous vehicles under adversarial attacks. In this article, we investigate the impact of two primary types of adversarial attacks, perturbation attacks, and patch attacks, on the driving safety of vision-based autonomous vehicles rather than the detection precision of deep learning models. In particular, we consider two state-of-the-art models in vision-based 3-D object detection: 1) Stereo R-CNN and 2) DSGN. To evaluate driving safety, we propose an end-to-end evaluation framework with a set of driving safety performance metrics. By analyzing the results of our extensive evaluation experiments, we find that: 1) the attack’s impact on the driving safety of autonomous vehicles and the attack’s impact on the precision of 3-D object detectors are decoupled and 2) the DSGN model demonstrates stronger robustness to adversarial attacks than the Stereo R-CNN model. In addition, we further investigate the causes behind the two findings with an ablation study. The findings of this article provide a new perspective to evaluate adversarial attacks and guide the selection of deep learning models in autonomous driving.

Network controllability measures how well a networked system can be controlled to a target state, and its robustness reflects how well the system can maintain the controllability against malicious attacks by means of node removals or edge removals. The measure of network controllability is quantified by the number of external control inputs needed to recover or to retain the controllability after the occurrence of an unexpected attack. The measure of the network controllability robustness, on the other hand, is quantified by a sequence of values that record the remaining controllability of the network after a sequence of attacks. Traditionally, the controllability robustness is determined by attack simulations, which is computationally time consuming. In this article, a method to predict the controllability robustness based on machine learning using a convolutional neural network (CNN) is proposed, motivated by the observations that: 1) there is no clear correlation between the topological features and the controllability robustness of a general network; 2) the adjacency matrix of a network can be regarded as a grayscale image; and 3) the CNN technique has proved successful in image processing without human intervention. Under the new framework, a fairly large number of training data generated by simulations are used to train a CNN for predicting the controllability robustness according to the input network-adjacency matrices, without performing conventional attack simulations. Extensive experimental studies were carried out, which demonstrate that the proposed framework for predicting controllability robustness of different network configurations is accurate and reliable with very low overheads.

Network controllability robustness (CR) reflects how well a networked system can maintain its controllability against destructive attacks. Its measure is quantified by a sequence of values that record the remaining controllability of the network after a sequence of node-removal or edge-removal attacks. Traditionally, the CR is determined by attack simulations, which is computationally time-consuming or even infeasible. In this article, an improved method for predicting the network CR is developed based on machine learning using a group of convolutional neural networks (CNNs). In this scheme, a number of training data generated by simulations are used to train the group of CNNs for classification and prediction, respectively. Extensive experimental studies are carried out, which demonstrate that 1) the proposed method predicts more precisely than the classical single-CNN predictor; 2) the proposed CNN-based predictor provides a better predictive measure than the traditional spectral measures and network heterogeneity.

Augmented reality overlays virtual objects on the real world. To do so, the head mounted display (HMD) needs to be calibrated to establish a mapping between 3D points in the real world with 2D pixels on display panels. This distortion is a high-dimensional function that also depends on pupil position and varifocal settings. We present Raycast calibration, an efficient approach to geometrically calibrate AR displays with off-axis reflective combiners. Our approach requires a small amount of data to estimate a compact, physics-based, and ray-traceable model of the HMD optics. We apply this technique to automatically calibrate an AR prototype with display, SLAM and eye-tracker, without user in the loop.

Restart strategy helps the covariance matrix adaptation evolution strategy (CMA-ES) to increase the probability of finding the global optimum in optimization, while a single run CMA-ES is easy to be trapped in local optima. In this paper, the continuous non-revisiting genetic algorithm (cNrGA) is used to help CMA-ES to achieve multiple restarts from different sub-regions of the search space. The CMA-ES with on-line search history-assisted restart strategy (HR-CMA-ES) is proposed. The entire on-line search history of cNrGA is stored in a binary space partitioning (BSP) tree, which is effective for performing local search. The frequently sampled sub-region is reflected by a deep position in the BSP tree. When leaf nodes are located deeper than a threshold, the corresponding sub-region is considered a region of interest (ROI). In HR-CMA-ES, cNrGA is responsible for global exploration and suggesting ROI for CMA-ES to perform an exploitation within or around the ROI. CMA-ES restarts independently in each suggested ROI. The non-revisiting mechanism of cNrGA avoids to suggest the same ROI for a second time. Experimental results on the CEC 2013 and 2017 benchmark suites show that HR-CMA-ES performs better than both CMA-ES and cNrGA. A positive synergy is observed by the memetic cooperation of the two algorithms.

To solve economic load dispatch problems, this paper designs a combination of artificial bee colony (ABC) and covariance matrix adaptation evolution strategy (CMA-ES). In this method, multiple variables are updated at the employed bee stage. The onlooker bee stage of the ABC method is replaced by the CMA-ES method. To begin with a good position, the CMA-ES method is initialized based on the state of employed bees of the ABC method. The proposed method is used to solve economic load dispatch problem with different sizes, and compared with three other methods. Simulation results show that the method attains better performance by combining ABC and CMA-ES. Moreover, the sensitivity of parameter settings is also discussed, and a default setting is obtained for such problems.

It is widely accepted that optimization of imaging system performance should be guided by task-based measures of image quality. It has been advocated that imaging hardware or data-acquisition designs should be optimized by use of an ideal observer that exploits full statistical knowledge of the measurement noise and class of objects to be imaged, without consideration of the reconstruction method. In practice, accurate and tractable models of the complete object statistics are often difficult to determine. Moreover, in imaging systems that employ compressive sensing concepts, imaging hardware and sparse image reconstruction are innately coupled technologies. In this paper, a sparsity-driven observer (SDO) that can be employed to optimize hardware by use of a stochastic object model describing object sparsity is described and investigated. The SDO and sparse reconstruction method can, therefore, be "matched" in the sense that they both utilize the same statistical information regarding the class of objects to be imaged. To efficiently compute the SDO test statistic, computational tools developed recently for variational Bayesian inference with sparse linear models are adopted. The use of the SDO to rank data-acquisition designs in a stylized example as motivated by magnetic resonance imaging is demonstrated. This paper reveals that the SDO can produce rankings that are consistent with visual assessments of the reconstructed images but different from those produced by use of the traditionally employed Hotelling observer.

The continuous Non-revisiting Genetic Algorithm (cNrGA) uses the entire search history and parameter-less adaptive mutation to significantly enhance search performance. Experimental results show that it has better performance than Covariance Matrix Adaptation Evolution Strategy (CMA-ES), a state of the art evolutionary algorithm. Storing the search history is natural and costs little when fitness evaluations are expensive. However, if the number of evaluations required is substantial, some memory management is desirable. In this paper, we propose two pruning mechanisms to keep the memory used constant. They are Random pruning and Least Recently Used pruning. The idea is to prune a node when a memory threshold is reached and a new node is required to be inserted, thus keeping the overall memory used constant. Experimental results show that both strategies can maintain the performance of cNrGA, up to the limit when 90% of the nodes are not recorded. This suggests that cNrGA can be extended to use in situations when the number of fitness evaluations are much larger than before with no significant effect on statistical performance, which widens the applicability of cNrGA to include more practical problems that require larger number of fitness evaluations before converging to the global optimum.

Evolutionary algorithms are typically run several times in design optimization problems and the best solution taken. We propose a novel online algorithm selection framework that learns to use the best algorithm based on previous runs, hence in effect using different and better algorithms as the search progresses. First, a set of algorithms are run on a benchmark problem suite. Given a new problem, a default algorithm is run and its convergence characteristics are recorded. This is used to map to the problem database to find the most similar problem. In turn, the database returns the best algorithm for this problem and this algorithm is run in the second iteration and so on, aiming to home onto the most suitable algorithm for the problem. The resulting algorithm, named Sequential Learnable Evolutionary algorithm (SLEA), outperforms Covariance Matrix Adaptation Evolution Strategy (CMA-ES) with multi-restarts. SLEA is also applied to a new problem, a real world application, and learns its characteristics. Experimental results show that it can correctly select the best algorithm for the problem. Finally, this paper proposes a new research program which learns the algorithm-problem mapping through solving real world problems accessed through the web and worldwide cooperation through Wikipedia.