Generative models have shown promising results in capturing human mobility characteristics and generating synthetic trajectories. However, it remains challenging to ensure that the generated geospatial mobility data is semantically realistic, including consistent location sequences, and reflects real-world characteristics, such as constraining on geospatial limits. We reformat human mobility modeling as an autoregressive generation task to address these issues, leveraging the Generative Pre-trained Transformer (GPT) architecture. To ensure its controllable generation to alleviate the above challenges, we propose a geospatially-aware generative model, MobilityGPT. We propose a gravity-based sampling method to train a transformer for semantic sequence similarity. Then, we constrained the training process via a road connectivity matrix that provides the connectivity of sequences in trajectory generation, thereby keeping generated trajectories in geospatial limits. Lastly, we proposed to construct a preference dataset for fine-tuning MobilityGPT via Reinforcement Learning from Trajectory Feedback (RLTF) mechanism, which minimizes the travel distance between training and the synthetically generated trajectories. Experiments on real-world datasets demonstrate MobilityGPT’s superior performance over state-of-the-art methods in generating high-quality mobility trajectories that are closest to real data in terms of origin-destination similarity, trip length, travel radius, link, and gravity distributions. We release the source code and reference links to datasets athttps://github.com/ammarhydr/MobilityGPT

Sonar imaging serves as a crucial sensing modality in underwater environments, yet its outputs are inherently de graded by substantial speckle noise arising from background in terference, biological activity, and device self-noise, posing unique challenges for effective denoising. Existing sonar image denoising techniques take a deterministic approach and produce only one outcome. However, this may neglect the complexities present in real-world situations, thereby limiting the effectiveness of denoising. To address this challenge, we first create a sonar noise dataset with corresponding reference images obtained through well esigned pairwise comparisons. Subsequently, we introduce the Divergence-to-Convergence Sonar Image Denoising (D2C SID) network. This network treats denoising as an uncertainty problem and generates multiple plausible denoised samples. It accomplishes this by leveraging a latent space constructed via a conditional variational autoencode. By this process, our method reduces the reliance on any single estimation. Experimental results on both forward-looking and side-scan sonar images confirm the efficacy of the D2C-SID approach.

The identification of drug-target interactions (DTIs) is a key step in drug discovery, but existing models lack sufficient generalization and robustness when processing data from diverse sources with inconsistent distributions. To address this problem, we propose MSCL-DTI: a novel dual-channel supervised contrastive learning framework for DTI prediction. The model integrates sequence-based and structure-based multi-source joint representations. These features are aggregated via a fusion module and jointly optimized under the guidance of a supervised contrastive learning objective, ultimately forming a unified and highly discriminative latent representation. Extensive experiments demonstrate that MSCL-DTI outperforms baseline models in both performance and generalization. Furthermore, the model was successfully applied to screen drugs for the SARS-CoV-2 Spike protein, proving it to be an effective tool for discovering potential targeted drugs. Data and code are available at https://github.com/HelloJie14492IMSCL-DTI.

This paper presents the sufficient conditions of the stability for discrete-time systems with random initial state shifts when applying iterative learning control. Different sufficient conditions have been developed for systems with different characteristics to ensure boundedness of tracking errors and convergence of the learning process under random initial condition. First, the relationship between the initial shifts of two adjacent iterations is addressed with the help of the transition matrix. And the different analysis models are established for each type of systems: one-dimensional matrix analysis model and 2-D Roesser model. Second, the new and sufficient conditions are established for the proposed learning control law to ensure systems convergence by matrix theory and two-dimensional system theory. Third, solving the linear matrix inequality (LMI) yields the control gains. Finally, digital simulations illustrate the effectiveness and sufficiency of the presented conditions.

Differentially-private (DP) Generative Adversarial Networks (GAN) can be used to protect the privacy of training data and support public downstream learning tasks with synthetic data. However, typical DP mechanisms add noise to the training process and can lead to various convergence problems. We propose HE-GAN, a DP generative framework that eliminates noise addition by using Exponential Mechanism (EM) on the privacy-factor-adjusted posterior predictive distribution of a classifier trained on the private data. EM is more general than many other DP mechanisms including Laplacian and Gaussian mechanisms. EM’s reliance on sampling the output space also prevents the DP noise from corrupting the training process. However, there are two challenges: first, sampling the posterior distribution of the private discriminative classifier may not be able to produce high-quality synthetic samples. Instead, we sample from the latent space of a publicly-trained GAN to optimize the private posterior. Second, we use the highly effective Hamiltonian Monte Carlo (HMC) method for latent space sampling. We perform experiments on MNIST and Fashion-MNIST under public-private splits. Results show that HE-GAN can achieve downstream classification accuracy on par with or better than state-of-the-art scheme over a wide range of privacy budgets.

Synchronizations are fundamental to the correctness and performance of concurrent software. Unfortunately, correctly identifying all synchronizations has become extremely difficult in modern soft-ware systems due to the various types of synchronizations. Previous work either only infers specific type of synchronization by code analysis or relies on manual effort to annotate the synchronization. This paper proposes SherLock, a tool that uses unsupervised inference to identify synchronizations. SherLock leverages the fact that most synchronizations appear around the conflicting operations and form it into a linear system with a set of synchronization proper-ties and hypotheses. To collect enough observations, SherLock runs the unit tests a small number of times with feedback-based delay injection. We applied SherLock on 8 C# open-source applications. Without any prior knowledge, SherLock inferred 122 unique synchronizations, with few false positives. These inferred synchronizations cover a wide variety of types, including lock operations, fork-join operations, asynchronous operations, framework synchronization, and custom synchronization.

Generating high-quality system call sequences is not only important to testing file system implementations, but also challenging due to the astronomically large input space. This paper introduces a new approach to the workload generation problem by building layered models and abstract workloads refinement. This approach is instantiated as a three-layer file system model for file system workload generation. In a short-period experiment run, sequential workloads (system call sequences) manifested over a thousand crashes in mainline Linux Kernel file systems, with 12 previously unknown bugs being reported. We also provide evidence that such workloads benefit other domain-specific testing techniques including crash consistency testing and concurrency testing.

A multithreaded program's interleaving space is discrete and astronomically large, making effectively sampling thread schedules for manifesting concurrency bugs a challenging task. Observing that concurrency bugs can be manifested by adjusting thread relative speeds, this paper presents the new concept of speed space in which each vector denotes a family of thread schedules. A multithreaded program's speed space is approximately continuous, easy-to-sample, and preserves certain categories of concurrency bugs. We discuss the design, implementation, and evaluation of our speed-controlled scheduler for exploring adversarial/abnormal schedules. The experimental results confirm that our technique is effective in sampling diverse schedules. Our implementation also found previously unknown concurrency bugs in real-world multithreaded programs.