Deep learning and computer vision have experienced significant advancements, particularly in critical applications such as autonomous driving and real-time surveillance, where object detection (OD) plays a pivotal role. Ensuring the accuracy and speed of these systems is paramount to prevent accidents or failures. Recently, latency-based attacks have emerged as a new threat, driven by the essential need for real-time performance in various applications. These attacks target model responsiveness to disrupt system performance without necessarily compromising accuracy. Our preliminary experiments show that introducing just a few bit flips to key parameters in OD models can significantly increase latency, degrading performance. Meanwhile, recent advancements in memory-based attacks, such as Row Hammer [18], demonstrate the ability to conveniently introduce bit flips at desired locations without physical hardware interaction. Based on the observations, we propose a novel attack on OD models that leverages row-hammer to introduce bit-flips via side channels, targeting the non-maximum suppression (NMS) filter and significantly increasing latency. Unlike previous methods that modify input data, our technique ensures efficiency by minimizing bit-flips through critical path exploitation and achieves practical applicability with only a subset of validation data. Experiments across various datasets and models validate our approach, demonstrating latency increases up to 71.6 ms (20.4×) with just 31 bit-flips.

Multi-user Augmented Reality (MuAR) allows multiple users to interact with shared virtual objects, facilitated by exchanging environment information. Current MuAR systems rely on 3D point clouds for real-world analysis, view synchronization, object rendering, and movement tracking. However, the complexity of 3D point clouds leads to significant processing delays, with approximately 80% of overhead in commercial frameworks. This hampers usability and degrades user experience. Our analysis reveals that maintaining the facing side of the real-world scene in a stable environment provides sufficient information for virtual object placement and rendering. To address this, we introduce a lightweight quadtree structure, representing 2D scenes through semantic segmentation and geometry, as an alternative to 3D point clouds. Additionally, we propose a novel correction method to handle potential shifts in virtual object placement during view synchronization among users. Combining all designs, we implement a fast and lightweight MuAR framework namedAR-Lightand test our framework on commercial AR devices. The evaluation results on real-world applications demonstrate that AR-Light can achieve high performance in various real-world scenes while maintaining a comparable virtual object placement accuracy.

Deep learning techniques have been used to enhance the joint inversion process of dc resistivity and seismic travel time data. Specifically, we introduce deep perceptual losses (DPLs) derived from a pretrained edge detection network. These losses play a pivotal role in enforcing structural constraints during the training of encoder-decoder networks for model prediction. Our approach is based on the assumption of structural similarity, meaning that we presume a shared structure between pairs of resistivity and velocity models, without necessitating prior relationships between the property values. We provide an exhaustive exposition of our joint inversion framework, elucidating the network construction and training dataset design. Numerical examples demonstrate our DPL-based method outperforms the conventional separate inversions and cross-gradient-based joint inversion in view of recovered property values and boundaries of the anomalous bodies. Furthermore, our approach retains the traditional workflows of separate inversions, endowing our framework with the flexibility to expand into multiphysics joint inversion scenarios. To illustrate this adaptability, we incorporate induced polarization data into the framework, further validating its efficacy.

In the context of Bayesian inversion for scientific and engineering modeling, Markov chain Monte Carlo (MCMC) sampling strategies have become the benchmark due to their flexibility and robustness in dealing with arbitrary posterior probability density functions (PDFs). However, these algorithms have been shown to be inefficient when sampling from high-dimensional posterior distributions or exhibit multimodality and/or strong parameter correlations. In such contexts, transitional MCMC (TMCMC) provides a more efficient alternative. Despite the recent applicability for Bayesian updating and model selection across a variety of disciplines, TMCMC may require a prohibitive number of tempering stages when the prior pdf is significantly different from the target posterior. Furthermore, the need to start with an initial set of samples from the prior distribution may present a challenge when dealing with implicit priors, e.g., based on feasible regions. Finally, TMCMC cannot be used for inverse problems with improper prior PDFs that represent a lack of prior knowledge on all or a subset of parameters. A generalization of TMCMC is proposed that alleviates such challenges and limitations toward providing a more robust and efficient tempering sampling strategy. We present convergence analysis, proving that the distance between the intermediate distributions and the target posterior distribution monotonically decreases as the algorithm proceeds. We also demonstrate the advantages of the proposed generalization through a series of test problems and an engineering application in the oil and gas industry.

In this abstract, we introduce a deep learning enhanced (DLE) framework for solving multi-physics joint inversion. For the inceptive DLE joint inversion scheme, a well-trained neural network based on structural similarity is used to provide better initials for the separate inversions. The preservation of the separate inversions makes the framework flexible to deal with different sensing configurations, nonconforming discretization, as well as joint inversion of multiple data types. Further, deep perceptual losses (DPL) derived from a pre-trained edge detection network are introduced to enforce structural constraints. Synthetic examples demonstrate improved inversion results from the DLE framework compared to the separate inversions and cross-gradient-based joint inversion. In addition, the DLE framework is simplified to solve 3D joint inversion of the airborne magnetic and gravity gradient data collected from the Decorah area. The inversion results verify the effectiveness and higher efficiency of the DL-based method compared to the cross-gradient-based joint inversion.

We introduce the Safe Offline-to-Online Multi-Agent Decision Transformer (SO2-MADT), an innovative framework that revolutionizes safety considerations in Multi-agent Reinforcement Learning (MARL) through a novel sequence modeling approach. Leveraging the dynamic capabilities inherent in Decision Transformers, our methodology seamlessly incorporates safety protocols as a cornerstone element, ensuring secure operations throughout both the offline pre-training phase and the adaptive online fine-tuning phase. At the core of our framework lie two pivotal innovations: the Safety-To-Go (STG) token, embedding safety at a macro level, and the Agent Prioritization Module (APM), facilitating explicit credit assignment at a micro level. Through extensive testing against the challenging environments of the StarCraft Multi-Agent Challenge (SMAC) and Multi-agent MuJoCo, our SO2-MADT not only excels in offline pre-training but also demonstrates superior performance during online fine-tuning, without any degradation in performance. The implications of our work provide a pathway for deployment in critical real-world applications where safety is paramount and non-negotiable. The code is available at https://github.com/shahaamirbader/SO2-MADT.

Seismic data interpolation is a vital technology for improving seismic data density. In recent years, deep learning approaches have demonstrated significant potential in this field, yielding impressive results. Nonetheless, challenges still persist and have not been adequately addressed. First, the lack of reliable labeled training datasets induces concerns about the network’s adaptiveness under supervised learning schemes. Additionally, due to inadequate spatial sampling, aliasing frequently poses considerable difficulties for deep neural networks. In this study, we tackle the issue of aliased seismic data interpolation through self-supervised learning. A novel dip-informed neural network (DINN) is introduced to explicitly integrate local dip information into the neural network and regularize the reconstruction of missing traces. To address the training challenges associated with regularly sampled seismic data interpolation under self-supervised learning schemes, a randomized mix training algorithm is developed. The experimental results along with comparisons to existing methods using both synthetic and field datasets demonstrate the effectiveness and robustness of our approach.

Accurately localizing 1D signal patterns, such as Gamma-ray well-log depth matching, is crucial in the oilfield service industry as it directly affects the quality of oil and gas exploration. However, traditional methods such as well-log curve analysis and pattern hand-picking matching are labor-intensive and heavily rely on human expertise, leading to inconsistent results. Although attempts have been made to automate this process, challenges such as low computational performance, non-robustness, and non-generalization remain unsolved. To address these challenges, we have developed a data-driven AI system that learns an active signal pattern localization strategy inspired by human attention. Our artificial intelligence system uses an offline reinforcement learning (RL) framework as its central component, which solves a highly abstracted Markov decision process problem via offline training on human-labeled historical data. The RL agent uses top-down reasoning to determine the location of target signal fragments by deforming a bounding window using simple transformation actions. To overcome distribution shifts between logged data and real and ensure generalization, we propose a discrete distributionally robust soft actor-critic RL framework (DRSAC-Discrete) to solve the Markov decision process problem under uncertainty. By exploring unfamiliar environments in a restrictive manner, the DRSAC-Discrete algorithm provides a safe solution that can be used when data is limited during the early stage of this industrial application. We evaluated the reinforcement learning-based localization system on augmented field Gamma-ray well-log datasets, and the results showed promising localization capability. Furthermore, the DRSAC-Discrete algorithm demonstrated relatively robust performance guarantees when facing data shortage.

The simultaneous-source technology for high-density seismic acquisition is a key solution to efficient seismic surveying. It is a cost-effective method when blended subsurface responses are recorded within a short time interval using multiple seismic sources. A following deblending process, however, is needed to separate signals contributed by individual sources. Recent advances in deep learning and its data-driven approach toward feature engineering have led to many new applications for a variety of seismic processing problems. It is still a challenge, though, to collect enough labeled data and avoid model overfitting and poor generalization performance over different datasets with a low resemblance from each other. In this article, we propose a novel self-supervised learning method to solve the deblending problem without labeled training datasets. Using a blind-trace deep neural network and a carefully crafted blending loss function, we demonstrate that the individual source-response pairs can be accurately separated under three different blended-acquisition designs.

Full-waveform inversion (FWI) plays a significant role in producing high-resolution subsurface imaging in seismic prospecting and ground penetrating radar (GPR). However, FWI faces various challenges in practice. For example, the lack of low-frequency information due to acquisition limitations will make the FWI prone to falling to the local minimum. In this project, a deep learning-based approach is proposed to extrapolate the low-frequency data. Specifically, we propose a robust progressive learning (RPL) algorithm that combines physics-guided FWI and data-driven deep learning technology. The proposed method is robust against the choice of the initial model. Experimental results show that our method can achieve high efficiency and accuracy by using a limited amount of training data. The subsurface structures are successfully reconstructed with our extrapolated low-frequency data.

Higher dimensional (i.e., 2-D and 3-D) modeling is indispensable to correctly evaluate the responses of electromagnetic (EM) logging tools in complex formation environments. However, limited by the high computational cost of rigorous modeling, such as the finite-difference method and the finite-element method, the real-time applications in the well logging industry primarily rely on the 1-D forward solver, which would result in erroneous formation evaluation for complex scenarios. As a result, aiming at realizing fast modeling for EM logging tools in complex formations, this letter proposes a general framework assisted by deep neural networks (DNNs). The framework consists of three modules: earth model classification, parameter extraction, and surrogate construction. Separate DNNs are trained and tested for different modules. The accuracy and efficiency of the DNN-assisted fast modeling are validated by several experiments. This study finds that the fast modeling assisted by DNNs is able to calculate the tool responses and reconstruct the subsurface formations in real time.

The low-frequency seismic data provide crucial information for guiding the full-waveform inversion (FWI), especially when strong reflectors exist in the velocity model. However, hardware limitations make it difficult to acquire low-frequency data. To overcome the nonlinearity and ill-posedness caused by the absence of the low-frequency data, we develop an efficient progressive transfer learning algorithm for low-frequency extrapolation. The proposed method combines the FWI, the sparsity-promoted bandwidth-extension (BWE) algorithm, and the physics-guided data-driven deep learning approach. Compared with pure data-driven learning-based methods and the original progressive transfer learning method without BWE, our proposed algorithm shows better generalization ability. By integrating the physics constraints and the BWE algorithm, the performance of our method is less dependent on the quality of the initial training velocity model and the corresponding training set. We propose a logarithmic transformation to rebalance the loss function to overcome the challenge of predicting the weak reflection low-frequency data. To accelerate the algorithm, we propose a learning-based BWE method for initializing the training set and a truncated FWI method to reduce the iterative workflow’s computational cost. Experimental results show that our method achieves both high efficiency and high accuracy. The subsurface structures below the strong reflectors are successfully reconstructed with our extrapolated low-frequency data.

Reconstruction of seismic data is an important but challenging task in seismic data processing. Different machine-learning-based algorithms have been developed to solve this ill-posed problem and achieved great progress. However, most machine-learning-based methods rely on supervised learning where a good training dataset with many complete shot-gathers are required to train the model. Although the generative model has been used for unsupervised learning and reconstructing signals in a shot-gather, it fails to accurately resolve the fine features, especially when aliasing is the main concern. In addition, multiple shots’ interpolation problems have not been fully investigated by the unsupervised machine-learning-based approaches. In this work, we propose a self-supervised learning method using a blind-trace network and two antialiasing techniques (automatic spectrum suppression and mix-training) for seismic data reconstruction. The method is validated using challenging and realistic scenarios. Test results show that the method can be applied to single-shot or multiple shots’ cases and adapt well to different decimation patterns.

In this article, a new scheme based on the supervised descent method (SDM) for solving directional electromagnetic logging-while-drilling (LWD) inverse problems is proposed. The SDM provides us a new perspective to combine the classical gradient-based inversion and machine-learning-based inversion schemes. It iteratively learns a set of descent directions in the offline training process, where the training model set is generated in advance according to the prior information, and then updates the models with the learned descent directions as well as data residuals in the prediction stage, resulting in great flexibility to incorporate prior information, the capability of skipping local minima, and accelerated convergence. The generalization ability of the SDM to interrogate new models that are not contained in the training model set is also explored. By utilizing real-time information obtained from the logging process, the learned descent directions can be slightly revised with a higher efficiency to get closer to the true model. In addition, we probe the sensitivity of the SDM by adding different levels of random noise to the measurements. Numerical examples demonstrate that SDM-based inversion can achieve a higher resolution, faster convergence, and higher robustness than conventional schemes such as Occam's inversion.

First break picking is an inevitable process in land seismic data processing, which involves a huge amount of human labor to perform. Even after decades of investigation on the first break picking process, there are still enormous challenges in developing a robust automatic approach. Although many experts proposed techniques to solve the first break picking problems automatically, there are no solid solutions to avoid human labors during the picking process. In the late 20th century, the rise of the artificial intelligence and the advancement of computer hardware have overcome some challenges in first break picking but the level of their success is limited. In this article, we proposed a deep machine learning model to achieve automatic seismic first break picking. Our proposed model can find the underlying factors and determine the first break curve. In addition, the network is capable of updating itself through continuous learning. The system is able to identify labeling anomalies on-site and update the model through active learning. Unfortunately, training the machine learning model on a huge data set that contains unnecessary data points is an inefficient way for both model learning process and human labeling labors. Therefore, training the model with data selected by the experts can highly reduce the training time and the number of data that human has to label. In simulation, we show the advantage of our proposed deep semisupervised neural network, which uses both labeled and unlabeled data sets to achieve higher accuracy compared with the supervised neural networks.

Although multi-view datasets have become more accessible in the real-world applications, most state-of-the-art action recognition methods applied to those datasets rely on simple view agreement when combining local information from various views together. This leads to deteriorated performance in situations with view insufficiency and view disagreements. In this paper, we propose a novel framework for boosting action recognition performance by quantifying the connection between the viewpoint and an action. The proposed approach searches for the best combination of multiple views based on a co-learning strategy that simultaneously learns a local distance metric related to each action class and the relationships between each viewpoint and the action category. Consequently, the spatio-temporal representation of each action class in different viewpoints plays a key role in shaping the local distance metric space. We test our method on the IXMAS dataset and shows competitive performance compared to other state-of-the-art methods.

Segmentation of cytological smears plays a critical role in the automated analysis of histological abnormalities by fine needle aspiration cytology. However, smears obtained from fine needle aspiration biopsy are often contaminated with blood. Segmentation of such an image is not a trivial task and the false positive rate could be high if the blood cells cannot be correctly separated from the rest of the sample. Moreover, the fine textured nature of the cell chromatin gives it a non-uniform intensity appearance in both color and gray images. In this paper, we propose an enhanced watershed approach to remove background noise by using short wavelength spectral image and the computed absorption image to improve segmentation accuracy. We also demonstrate a color image segmentation method by applying watershed to the minima imposed aggregation image. Results of segmentation on 20 images of cytological smears are presented and the accuracy compared for the two methods.