GPU-embedded systems have gained popularity across various domains due to their efficient power consumption. However, in order to meet the demands of real-time or time-consuming applications running on these systems, it is crucial for them to be tuned to exhibit high performance. This paper addresses the issue by developing and comparing two tuning methodologies on GPU-embedded systems, and also provides performance insights for developers and researchers seeking to optimize applications running on these architectures. We focus on parallel prefix operations, such as FFT, scan primitives, and tridiagonal system solvers, which are performance-critical components in many applications. The study introduces an analytical model-driven tuning methodology and a Machine Learning (ML)-based tuning methodology. We evaluate the performance of the two tuning methodologies for different parallel prefix implementations of the BPLG library in an NVIDIA Jetson system, and compare their performance to the ones achieved through an exhaustive search. The findings shed light on the best strategies for handling the open challenge of performance portability for major computational patterns among server and embedded devices, providing practical guidance for offline and online tuning. We also address the existing gap in performance studies for parallel computational patterns in GPU-embedded systems by comparing the BPLG performance against other state-of-the-art libraries, including CUSPARSE, CUB, and CUFFT.
相關內容
Wayfinding in complex indoor environments is often challenging for older adults due to declines in navigational and spatial-cognition abilities. This paper introduces NavMarkAR, an augmented reality navigation system designed for smart-glasses to provide landmark-based guidance, aiming to enhance older adults' spatial navigation skills. This work addresses a significant gap in design research, with limited prior studies evaluating cognitive impacts of AR navigation systems. An initial usability test involved 6 participants, leading to prototype refinements, followed by a comprehensive study with 32 participants in a university setting. Results indicate improved wayfinding efficiency and cognitive map accuracy when using NavMarkAR. Future research will explore long-term cognitive skill retention with such navigational aids.
Unmanned aerial vehicles (UAVs) have gained popularity in the communications research community because of their versatility in placement and potential to extend the functions of communication networks. However, there remains still a gap in existing works regarding detailed and measurement-verified air-to-ground (A2G) Massive Multi-Input Multi-Output (MaMIMO) channel characteristics which play an important role in realistic deployment. In this paper, we first design a UAV MaMIMO communication platform for channel acquisition. We then use the testbed to measure uplink Channel State Information (CSI) between a rotary-wing drone and a 64-element MaMIMO base station (BS). For characterization, we focus on multidimensional channel stationarity which is a fundamental metric in communication systems. Afterward, we present measurement results and analyze the channel statistics based on power delay profiles (PDPs) considering space, time, and frequency domains. We propose the stationary angle (SA) as a supplementary metric of stationary distance (SD) in the time domain. We analyze the coherence bandwidth and RMS delay spread for frequency stationarity. Finally, spatial correlations between elements are analyzed to indicate the spatial stationarity of the array. The space-time-frequency channel stationary characterization will benefit the physical layer design of MaMIMO-UAV communications.
Quantifying variable importance is essential for answering high-stakes questions in fields like genetics, public policy, and medicine. Current methods generally calculate variable importance for a given model trained on a given dataset. However, for a given dataset, there may be many models that explain the target outcome equally well; without accounting for all possible explanations, different researchers may arrive at many conflicting yet equally valid conclusions given the same data. Additionally, even when accounting for all possible explanations for a given dataset, these insights may not generalize because not all good explanations are stable across reasonable data perturbations. We propose a new variable importance framework that quantifies the importance of a variable across the set of all good models and is stable across the data distribution. Our framework is extremely flexible and can be integrated with most existing model classes and global variable importance metrics. We demonstrate through experiments that our framework recovers variable importance rankings for complex simulation setups where other methods fail. Further, we show that our framework accurately estimates the true importance of a variable for the underlying data distribution. We provide theoretical guarantees on the consistency and finite sample error rates for our estimator. Finally, we demonstrate its utility with a real-world case study exploring which genes are important for predicting HIV load in persons with HIV, highlighting an important gene that has not previously been studied in connection with HIV. Code is available at //github.com/jdonnelly36/Rashomon_Importance_Distribution.
The pervasive deployment of surveillance cameras produces a massive volume of data, requiring nuanced interpretation. This study thoroughly examines data representation and visualization techniques tailored for AI surveillance data within current infrastructures. It delves into essential data metrics, methods for situational awareness, and various visualization techniques, highlighting their potential to enhance safety and guide urban development. This study is built upon real-world research conducted in a community college environment, utilizing eight cameras over eight days. This study presents tools like the Occupancy Indicator, Statistical Anomaly Detection, Bird's Eye View, and Heatmaps to elucidate pedestrian behaviors, surveillance, and public safety. Given the intricate data from smart video surveillance, such as bounding boxes and segmented images, we aim to convert these computer vision results into intuitive visualizations and actionable insights for stakeholders, including law enforcement, urban planners, and social scientists. The results emphasize the crucial impact of visualizing AI surveillance data on emergency handling, public health protocols, crowd control, resource distribution, predictive modeling, city planning, and informed decision-making.
The surge in electricity use, coupled with the dependency on intermittent renewable energy sources, poses significant hurdles to effectively managing power grids, particularly during times of peak demand. Demand Response programs and energy conservation measures are essential to operate energy grids while ensuring a responsible use of our resources This research combines distributed optimization using ADMM with Deep Learning models to plan indoor temperature setpoints effectively. A two-layer hierarchical structure is used, with a central building coordinator at the upper layer and local controllers at the thermal zone layer. The coordinator must limit the building's maximum power by translating the building's total power to local power targets for each zone. Local controllers can modify the temperature setpoints to meet the local power targets. The resulting control algorithm, called Distributed Planning Networks, is designed to be both adaptable and scalable to many types of buildings, tackling two of the main challenges in the development of such systems. The proposed approach is tested on an 18-zone building modeled in EnergyPlus. The algorithm successfully manages Demand Response peak events.
Vision Transformers (ViTs) achieve superior performance on various tasks compared to convolutional neural networks (CNNs), but ViTs are also vulnerable to adversarial attacks. Adversarial training is one of the most successful methods to build robust CNN models. Thus, recent works explored new methodologies for adversarial training of ViTs based on the differences between ViTs and CNNs, such as better training strategies, preventing attention from focusing on a single block, or discarding low-attention embeddings. However, these methods still follow the design of traditional supervised adversarial training, limiting the potential of adversarial training on ViTs. This paper proposes a novel defense method, MIMIR, which aims to build a different adversarial training methodology by utilizing Masked Image Modeling at pre-training. We create an autoencoder that accepts adversarial examples as input but takes the clean examples as the modeling target. Then, we create a mutual information (MI) penalty following the idea of the Information Bottleneck. Among the two information source inputs and corresponding adversarial perturbation, the perturbation information is eliminated due to the constraint of the modeling target. Next, we provide a theoretical analysis of MIMIR using the bounds of the MI penalty. We also design two adaptive attacks when the adversary is aware of the MIMIR defense and show that MIMIR still performs well. The experimental results show that MIMIR improves (natural and adversarial) accuracy on average by 4.19\% on CIFAR-10 and 5.52\% on ImageNet-1K, compared to baselines. On Tiny-ImageNet, we obtained improved natural accuracy of 2.99\% on average and comparable adversarial accuracy. Our code and trained models are publicly available\footnote{\url{//anonymous.4open.science/r/MIMIR-5444/README.md}}.
Massive MIMO systems are typically designed assuming linear power amplifiers (PAs). However, PAs are most energy efficient close to saturation, where non-linear distortion arises. For conventional precoders, this distortion can coherently combine at user locations, limiting performance. We propose a graph neural network (GNN) to learn a mapping between channel and precoding matrices, which maximizes the sum rate affected by non-linear distortion, using a high-order polynomial PA model. In the distortion-limited regime, this GNN-based precoder outperforms zero forcing (ZF), ZF plus digital pre-distortion (DPD) and the distortion-aware beamforming (DAB) precoder from the state-of-the-art. At an input back-off of -3 dB the proposed precoder compared to ZF increases the sum rate by 8.60 and 8.84 bits/channel use for two and four users respectively. Radiation patterns show that these gains are achieved by transmitting the non-linear distortion in non-user directions. In the four user-case, for a fixed sum rate, the total consumed power (PA and processing) of the GNN precoder is 3.24 and 1.44 times lower compared to ZF and ZF plus DPD respectively. A complexity analysis shows six orders of magnitude reduction compared to DAB precoding. This opens perspectives to operate PAs closer to saturation, which drastically increases their energy efficiency.
This article presents the affordances that Generative Artificial Intelligence can have in disinformation context, one of the major threats to our digitalized society. We present a research framework to generate customized agent-based social networks for disinformation simulations that would enable understanding and evaluation of the phenomena whilst discussing open challenges.
Unsupervised domain adaptation has recently emerged as an effective paradigm for generalizing deep neural networks to new target domains. However, there is still enormous potential to be tapped to reach the fully supervised performance. In this paper, we present a novel active learning strategy to assist knowledge transfer in the target domain, dubbed active domain adaptation. We start from an observation that energy-based models exhibit free energy biases when training (source) and test (target) data come from different distributions. Inspired by this inherent mechanism, we empirically reveal that a simple yet efficient energy-based sampling strategy sheds light on selecting the most valuable target samples than existing approaches requiring particular architectures or computation of the distances. Our algorithm, Energy-based Active Domain Adaptation (EADA), queries groups of targe data that incorporate both domain characteristic and instance uncertainty into every selection round. Meanwhile, by aligning the free energy of target data compact around the source domain via a regularization term, domain gap can be implicitly diminished. Through extensive experiments, we show that EADA surpasses state-of-the-art methods on well-known challenging benchmarks with substantial improvements, making it a useful option in the open world. Code is available at //github.com/BIT-DA/EADA.
Object detectors usually achieve promising results with the supervision of complete instance annotations. However, their performance is far from satisfactory with sparse instance annotations. Most existing methods for sparsely annotated object detection either re-weight the loss of hard negative samples or convert the unlabeled instances into ignored regions to reduce the interference of false negatives. We argue that these strategies are insufficient since they can at most alleviate the negative effect caused by missing annotations. In this paper, we propose a simple but effective mechanism, called Co-mining, for sparsely annotated object detection. In our Co-mining, two branches of a Siamese network predict the pseudo-label sets for each other. To enhance multi-view learning and better mine unlabeled instances, the original image and corresponding augmented image are used as the inputs of two branches of the Siamese network, respectively. Co-mining can serve as a general training mechanism applied to most of modern object detectors. Experiments are performed on MS COCO dataset with three different sparsely annotated settings using two typical frameworks: anchor-based detector RetinaNet and anchor-free detector FCOS. Experimental results show that our Co-mining with RetinaNet achieves 1.4%~2.1% improvements compared with different baselines and surpasses existing methods under the same sparsely annotated setting.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191