Improving Fast Adversarial Training With Prior-Guided Knowledge.

Fast adversarial training (FAT) is an efficient method to improve robustness in white-box attack scenarios. However, the original FAT suffers from catastrophic overfitting, which dramatically and suddenly reduces robustness after a few training epochs. Although various FAT variants have been proposed to prevent overfitting, they require high training time. In this paper, we investigate the relationship between adversarial example quality and catastrophic overfitting by comparing the training processes of standard adversarial training and FAT. We find that catastrophic overfitting occurs when the attack success rate of adversarial examples becomes worse. Based on this observation, we propose a positive prior-guided adversarial initialization to prevent overfitting by improving adversarial example quality without extra training time. This initialization is generated by using high-quality adversarial perturbations from the historical training process. We provide theoretical analysis for the proposed initialization and propose a prior-guided regularization method that boosts the smoothness of the loss function. Additionally, we design a prior-guided ensemble FAT method that averages the different model weights of historical models using different decay rates. Our proposed method, called FGSM-PGK, assembles the prior-guided knowledge, i.e., the prior-guided initialization and model weights, acquired during the historical training process. The proposed method can effectively improve the model's adversarial robustness in white-box attack scenarios. Evaluations of four datasets demonstrate the superiority of the proposed method.

Effects of tai chi, ba duan jin, and walking on the mental health status of urban older people living alone: the mediating role of social participation and the moderating role of the exercise environment.

With the global trend of aging, lacking of interpersonal communication and spiritual support and companionship have had a great impact on the mental health of older people living alone. This study examines the multifaceted impacts of engaging in tai chi, ba duan jin, and walking on the mental wellbeing of older people residing alone in urban areas. Additionally, this research aims to explore the association between tai chi, ba duan jin, and walking, and the mental health status of urban older people living alone, by considering the mediating influence of social participation and the moderating influence of the exercise environment. To do so, 1,027 older people living alone in six Chinese cities were investigated using the Physical Activity Rating Scale (PARS-3), the Geriatric Health Questionnaire (GHQ-12), the Social Participation Indicator System Scale, and the Exercise Environment Scale. SPSS 25.0 was utilized for conducting mathematical statistical analysis, specifically for doing linear regression analysis. Additionally, AMOS was employed to develop the study model. We found that a significant negative correlation between tai chi, ba duan jin, and walking and mental health status; among these, tai chi had the greatest impact on the mental health status of urban older people living alone. Social participation mediated the relationship between tai chi, ba duan jin, walking, and mental health status, and the exercise environment had a moderating effect on this relationship. The findings of this study indicate that tai chi, ba duan jin, and walking have a positive impact on the mental health of urban older people living alone, which can be influenced by the mediating efficacy of social participation and the moderating effect of the exercise environment.

Generalizable Black-Box Adversarial Attack With Meta Learning.

In the scenario of black-box adversarial attack, the target model's parameters are unknown, and the attacker aims to find a successful adversarial perturbation based on query feedback under a query budget. Due to the limited feedback information, existing query-based black-box attack methods often require many queries for attacking each benign example. To reduce query cost, we propose to utilize the feedback information across historical attacks, dubbed example-level adversarial transferability. Specifically, by treating the attack on each benign example as one task, we develop a meta-learning framework by training a meta generator to produce perturbations conditioned on benign examples. When attacking a new benign example, the meta generator can be quickly fine-tuned based on the feedback information of the new task as well as a few historical attacks to produce effective perturbations. Moreover, since the meta-train procedure consumes many queries to learn a generalizable generator, we utilize model-level adversarial transferability to train the meta generator on a white-box surrogate model, then transfer it to help the attack against the target model. The proposed framework with the two types of adversarial transferability can be naturally combined with any off-the-shelf query-based attack methods to boost their performance, which is verified by extensive experiments. The source code is available at https://github.com/SCLBD/MCG-Blackbox.

The Study of COVID-19's Perceptions as a Motivation to Exercise and Keep Fit Among Older Adults.

Since the end of 2019, a novel coronavirus pandemic (COVID-19), characterized by solid infectivity, rapid communication and diverse communication routes, has become widespread worldwide. This study investigates the motivations of older adults to exercise and keep fit due to the COVID-19 pandemic. The research is based on a survey of older adults in Chengdu, Sichuan Province, China. It adopts the event strength system theory as a conceptual framework and the stimulus-organism-response (S-O-R) theory for causal inferences. (1) the perception of COVID-19's novelty had a significant negative impact on older adults" intention to exercise and stay fit, and the perception of COVID-19's disruptiveness and criticality had a significant positive impact on older adults" intention to exercise and fitness; (2) The perception of COVID-19's novelty had a significant negative effect on risk cognition, and the disruptive and critical cognitions had a significant positive effect on risk perceptions; (3) risk perceptions had a prominent positive effect on older adults" intention to exercise and fitness; and (4) risk communication played an important moderating role between COVID-19 event strength cognition and older adults' intention to exercise and fitness. The study revealed that the perception of COVID-19's seriousness significantly impacted older adults" intentions to exercise and keep fit and that risk perception and communication acted as mediating factors.

Versatile Weight Attack via Flipping Limited Bits.

Many attack paradigms against deep neural networks have been well studied, such as the backdoor attack in the training stage and the adversarial attack in the inference stage. In this article, we study a novel attack paradigm, the bit-flip based weight attack, which directly modifies weight bits of the attacked model in the deployment stage. To meet various attack scenarios, we propose a general formulation including terms to achieve effectiveness and stealthiness goals and a constraint on the number of bit-flips. Furthermore, benefitting from this extensible and flexible formulation, we present two cases with different malicious purposes, i.e., single sample attack (SSA) and triggered samples attack (TSA). SSA which aims at misclassifying a specific sample into a target class is a binary optimization with determining the state of the binary bits (0 or 1); TSA which is to misclassify the samples embedded with a specific trigger is a mixed integer programming (MIP) with flipped bits and a learnable trigger. Utilizing the latest technique in integer programming, we equivalently reformulate them as continuous optimization problems, whose approximate solutions can be effectively and efficiently obtained by the alternating direction method of multipliers (ADMM) method. Extensive experiments demonstrate the superiority of our methods.

The influence of tai chi on the death anxiety of elderly people living alone: the chain mediating effect of social support and psychological capital.

Background: Population aging is a global trend, and the number of older adults living alone is increasing. Tai chi, a traditional Chinese exercise, has been shown to improve the physical and mental health of older adults. Aim: To investigate the effects of tai chi on death anxiety in older adults living alone and the role of social support and psychological capital in this relationship. Method: A cross-sectional study of 493 older adults living alone in four cities in southwestern China. Participants were assessed using questionnaires on tai chi practice, social support, psychological capital, and death anxiety. Results: Tai chi practice significantly reduced death anxiety in older adults living alone. It also positively correlated with social support and psychological capital, both of which negatively correlated with death anxiety. Social support and psychological capital mediated the relationship between tai chi practice and death anxiety, suggesting that tai chi may reduce death anxiety through these factors. These findings encourage older adults living alone to practice tai chi, as it may improve their mental and physical health and reduce their risk of death anxiety. Conclusion: Tai chi practice may reduce death anxiety in older adults living alone through the chain-mediated effects of social support and psychological capital. This suggests that tai chi may be a beneficial intervention for older adults living alone.

Boosting Fast Adversarial Training With Learnable Adversarial Initialization.

Adversarial training (AT) has been demonstrated to be effective in improving model robustness by leveraging adversarial examples for training. However, most AT methods are in face of expensive time and computational cost for calculating gradients at multiple steps in generating adversarial examples. To boost training efficiency, fast gradient sign method (FGSM) is adopted in fast AT methods by calculating gradient only once. Unfortunately, the robustness is far from satisfactory. One reason may arise from the initialization fashion. Existing fast AT generally uses a random sample-agnostic initialization, which facilitates the efficiency yet hinders a further robustness improvement. Up to now, the initialization in fast AT is still not extensively explored. In this paper, focusing on image classification, we boost fast AT with a sample-dependent adversarial initialization, i.e., an output from a generative network conditioned on a benign image and its gradient information from the target network. As the generative network and the target network are optimized jointly in the training phase, the former can adaptively generate an effective initialization with respect to the latter, which motivates gradually improved robustness. Experimental evaluations on four benchmark databases demonstrate the superiority of our proposed method over state-of-the-art fast AT methods, as well as comparable robustness to advanced multi-step AT methods. The code is released at https://github.com//jiaxiaojunQAQ//FGSM-SDI.

lp-Box ADMM: A Versatile Framework for Integer Programming.

This paper revisits the integer programming (IP) problem, which plays a fundamental role in many computer vision and machine learning applications. The literature abounds with many seminal works that address this problem, some focusing on continuous approaches (e.g., linear program relaxation), while others on discrete ones (e.g., min-cut). However, since many of these methods are designed to solve specific IP forms, they cannot adequately satisfy the simultaneous requirements of accuracy, feasibility, and scalability. To this end, we propose a novel and versatile framework called $\ell _p$ℓp-box ADMM, which is based on two main ideas. (1) The discrete constraint is equivalently replaced by the intersection of a box and an $\ell _p$ℓp-norm sphere. (2) We infuse this equivalence into the Alternating Direction Method of Multipliers (ADMM) framework to handle the continuous constraints separately and to harness its attractive properties. More importantly, the ADMM update steps can lead to manageable sub-problems in the continuous domain. To demonstrate its efficacy, we apply it to an optimization form that occurs often in computer vision and machine learning, namely binary quadratic programming (BQP). In this case, the ADMM steps are simple, computationally efficient. Moreover, we present the theoretic analysis about the global convergence of the $\ell _p$ℓp-box ADMM through adding a perturbation with the sufficiently small factor $\epsilon$ε to the original IP problem. Specifically, the globally converged solution generated by $\ell _p$ℓp-box ADMM for the perturbed IP problem will be close to the stationary and feasible point of the original IP problem within $O(\epsilon)$O(ε). We demonstrate the applicability of $\ell _p$ℓp-box ADMM on three important applications: MRF energy minimization, graph matching, and clustering. Results clearly show that it significantly outperforms existing generic IP solvers both in runtime and objective. It also achieves very competitive performance to state-of-the-art methods designed specifically for these applications.