Generating adversarial examples (AEs) can be formulated as an optimization problem. Among various optimization-based attacks, the gradient-based PGD and the momentum-based MI-FGSM have garnered considerable interest. However, all these attacks use the sign function to scale their perturbations, which raises several theoretical concerns from the point of view of optimization. In this paper, we first reveal that PGD is actually a specific reformulation of the projected gradient method using only the current gradient to determine its step-size. Further, we show that when we utilize a conventional adaptive matrix with the accumulated gradients to scale the perturbation, PGD becomes AdaGrad. Motivated by this analysis, we present a novel momentum-based attack AdaMI, in which the perturbation is optimized with an interesting momentum-based adaptive matrix. AdaMI is proved to attain optimal convergence for convex problems, indicating that it addresses the non-convergence issue of MI-FGSM, thereby ensuring stability of the optimization process. The experiments demonstrate that the proposed momentum-based adaptive matrix can serve as a general and effective technique to boost adversarial transferability over the state-of-the-art methods across different networks while maintaining better stability and imperceptibility.

Traffic flow prediction has emerged as a critical component in advancing smart cities. Nevertheless, precisely forecasting traffic flow remains a formidable challenge, attributable to the intricate and dynamic spatiotemporal interdependencies inherent within traffic data. Contemporary methodologies often overlook the different impacts exerted at each timestamp and treat temporal correlations, rendering them incapable of extracting temporal patterns at multiple scales. Furthermore, these approaches fail to account for the neighboring and functional relationships among nodes within the spatial module. In this work, we introduce an innovative Multiscale and Gated transFormer (MSGFormer) architecture, MSGFormer, to overcome the inherent limitations for accurate traffic flow estimation. We have introduced a multiscale sampling strategy wherein we sample from the original data at three scales: 1) recent; 2) daily; and 3) weekly. This approach enables the generation of corresponding subsequences that encapsulate temporal information across different granularities. Subsequently, each generated subsequence is projected into a latent space and systematically combined with positional, temporal, and spatial embeddings. The positional embedding comprises relative positional embedding and time stamp embedding corresponding to days and weeks, aiming at capturing the sequential and cyclical characteristics of the data. Furthermore, at the inception of the transformer encoder, a gated unit, composed of a neighboring mask and a functioning mask, is employed to capture both static and dynamic spatial correlations effectively. Comprehensive experiments have been conducted on four real-world traffic data sets. The experimental results robustly validate that our model attains significantly higher predictive accuracy in comparison to other baseline models.

The adversarial example presents new security threats to trustworthy detection systems. In the context of evading dynamic detection based on API call sequences, a practical approach involves inserting perturbing API calls to modify these sequences. The type of inserted API calls and their insertion locations are crucial for generating an effective adversarial API call sequence. Existing methods either optimize the inserted API calls while neglecting the insertion positions or treat these optimizations as separate processes. This can lead to inefficient attacks that insert a large number of unnecessary API calls. To address this issue, we propose a novel reinforcement learning (RL) framework, dubbed GAME-RL, which simultaneously optimizes both the perturbing APIs and their insertion positions. Specifically, we define malware modification through IAT (Import Address Table) hooking as a sequential decision-making process. We introduce an invalid action masking and an auto-regressive policy head within the RL framework, ensuring the feasibility of IAT hooking and capturing the inherent relationship between factors. GAME-RL learns more effective evasion strategies, taking into account functionality preservation and the black-box setting. We conduct comprehensive experiments on various target models, demonstrating that GAME-RL significantly improves the evasion rate while maintaining acceptable levels of adversarial overhead.

Averaging scheme has attracted extensive attention in deep learning as well as traditional machine learning. It achieves theoretically optimal convergence and also improves the empirical model performance. However, there is still a lack of sufficient convergence analysis for strongly convex optimization. Typically, the convergence about the last iterate of gradient descent methods, which is referred to as individual convergence, fails to attain its optimality due to the existence of logarithmic factor. In order to remove this factor, we first develop gradient descent averaging (GDA), which is a general projection-based dual averaging algorithm in the strongly convex setting. We further present primal-dual averaging for strongly convex cases (SC-PDA), where primal and dual averaging schemes are simultaneously utilized. We prove that GDA yields the optimal convergence rate in terms of output averaging, while SC-PDA derives the optimal individual convergence. Several experiments on SVMs and deep learning models validate the correctness of theoretical analysis and effectiveness of algorithms.