Search

The security of models based on new architectures such as MLP-Mixer and ViTs needs to be studied urgently. However, most of the current researches are mainly aimed at the adversarial attack against ViTs, and there is still relatively little adversarial work on MLP-mixer. We propose an adversarial attack method against MLP-Mixer called Maxwell's demon Attack (MA). MA breaks the channel-mixing and token-mixing mechanism of MLP-Mixer by controlling the part input of MLP-Mixer's each Mixer layer, and disturbs MLP-Mixer to obtain the main information of images. Our method can mask the part input of the Mixer layer, avoid overfitting of the adversarial examples to the source model, and improve the transferability of cross-architecture. Extensive experimental evaluation demonstrates the effectiveness and superior performance of the proposed MA. Our method can be easily combined with existing methods and can improve the transferability by up to 38.0% on MLP-based ResMLP. Adversarial examples produced by our method on MLP-Mixer are able to exceed the transferability of adversarial examples produced using DenseNet against CNNs. To the best of our knowledge, we are the first work to study adversarial transferability of MLP-Mixer.

The security of artificial intelligence (AI) is an important research area towards safe, reliable, and trustworthy AI systems. To accelerate the research on AI security, the Artificial Intelligence Security Competition (AISC) was organized by the Zhongguancun Laboratory, China Industrial Control Systems Cyber Emergency Response Team, Institute for Artificial Intelligence, Tsinghua University, and RealAI as part of the Zhongguancun International Frontier Technology Innovation Competition (https://www.zgc-aisc.com/en). The competition consists of three tracks, including Deepfake Security Competition, Autonomous Driving Security Competition, and Face Recognition Security Competition. This report will introduce the competition rules of these three tracks and the solutions of top-ranking teams in each track., Comment: Technical report of AISC

Additional file 1: Table S1. Sequence of the real-time PCR primers.

Since the micellization kinetics is influenced by polymer structure, the spherical three-dimensional topology of amphiphilic dendritic copolymers (ADPs) which hinders the phase separation during micellization is assumed to make the micellization kinetics different. In the literatures, most of the attention has been paid to the morphology transition or the morphology at equilibrium and the micellization kinetics of ADPs is rarely reported. In this study, the micellization processes of amphiphilic dendritic copolymers from unimers to the final equilibrium micelles were monitored by laser light scattering. Based on the closed association mechanism, the thermodynamics of micellization was analysed. The negative thermodynamic quantities indicate that the micellization of ADPs is driven by enthalpy. Based on the change of scattering intensity and hydrodynamic radius (Rh) with time, the detailed micellization kinetics was analysed, which contains two steps. By controlling the temperature and type of solvent, a system in which the concentration has little influence on Rh is obtained. The relaxation times of the two steps decrease with concentration, indicating that at higher concentration the rate of micellization is quicker. With the increasing mass fraction of the hydrophobic part, the relaxation times decrease and the driving force of micellization increases.