Neural Networks最新文献

Vehicular edge computing (VEC), a promising paradigm for the development of emerging intelligent transportation systems, can provide lower service latency for vehicular applications. However, it is still a challenge to fulfill the requirements of such applications with stringent latency requirements in the VEC system with limited resources. In addition, existing methods focus on handling the offloading task in a certain time slot with statically allocated resources, but ignore the heterogeneous tasks' different resource requirements, resulting in resource wastage. To solve the real-time task offloading and heterogeneous resource allocation problem in VEC system, we propose a decentralized solution based on the attention mechanism and recurrent neural networks (RNN) with a multi-agent distributed deep deterministic policy gradient (AR-MAD4PG). First, to address the partial observability of agents, we construct a shared agent graph and propose a periodic communication mechanism that enables edge nodes to aggregate information from other edge nodes. Second, to help agents better understand the current system state, we design an RNN-based feature extraction network to capture the historical state and resource allocation information of the VEC system. Thirdly, to tackle the challenges of excessive joint observation-action space and ineffective information interference, we adopt the multi-head attention mechanism to compress the dimension of the observation-action space of agents. Finally, we build a simulation model based on the actual vehicle trajectories, and the experimental results show that our proposed method outperforms the existing approaches.

Multi-focus image fusion (MFIF) is an important technique that aims to combine the focused regions of multiple source images into a fully clear image. Decision-map methods are widely used in MFIF to maximize the preservation of information from the source images. While many decision-map methods have been proposed, they often struggle with difficulties in determining focus and non-focus boundaries, further affecting the quality of the fused images. Dynamic threshold neural P (DTNP) systems are computational models inspired by biological spiking neurons, featuring dynamic threshold and spiking mechanisms to better distinguish focused and unfocused regions for decision map generation. However, original DTNP systems require manual parameter configuration and have only one stimulus. Therefore, they are not suitable to be used directly for generating high-precision decision maps. To overcome these limitations, we propose a variant called parameter adaptive dual channel DTNP (PADCDTNP) systems. Inspired by the spiking mechanisms of PADCDTNP systems, we further develop a new MFIF method. As a new neural model, PADCDTNP systems adaptively estimate parameters according to multiple external inputs to produce decision maps with robust boundaries, resulting in high-quality fusion results. Comprehensive experiments on the Lytro/MFFW/MFI-WHU dataset show that our method achieves advanced performance and yields comparable results to the fourteen representative MFIF methods. In addition, compared to the standard DTNP systems, PADCDTNP systems improve the fusion performance and fusion efficiency on the three datasets by 5.69% and 86.03%, respectively. The codes for both the proposed method and the comparison methods are released at https://github.com/MorvanLi/MFIF-PADCDTNP.

Real-world adversarial patches were shown to be successful in compromising state-of-the-art models in various computer vision applications. Most existing defenses rely on analyzing input or feature level gradients to detect the patch. However, these methods have been compromised by recent GAN-based attacks that generate naturalistic patches. In this paper, we propose a new perspective to defend against adversarial patches based on the entropy carried by the input, rather than on its saliency. We present Jedi, a new defense against adversarial patches that tackles the patch localization problem from an information theory perspective; leveraging the high entropy of adversarial patches to identify potential patch zones, and using an autoencoder to complete patch regions from high entropy kernels. Jedi achieves high-precision adversarial patch localization and removal, detecting on average 90% of adversarial patches across different benchmarks, and recovering up to 94% of successful patch attacks. Since Jedi relies on an input entropy analysis, it is model-agnostic, and can be applied to off-the-shelf models without changes to the training or inference of the models. Moreover, we propose a comprehensive qualitative analysis that investigates the cases where Jedi fails, comparatively with related methods. Interestingly, we find a significant core failure cases among the different defenses share one common property: high entropy. We think that this work offers a new perspective to understand the adversarial effect under physical-world settings. We also leverage these findings to enhance Jedi's handling of entropy outliers by introducing Adaptive Jedi, which boosts performance by up to 9% in challenging images.