Adversarial Machine Learning in Image Classification: A Survey Toward the Defender’s Perspective

Deep Learning algorithms have achieved state-of-the-art performance for Image Classification. For this reason, they have been used even in security-critical applications, such as biometric recognition systems and self-driving cars. However, recent works have shown those algorithms, which can even surpass human capabilities, are vulnerable to adversarial examples. In Computer Vision, adversarial examples are images containing subtle perturbations generated by malicious optimization algorithms to fool classifiers. As an attempt to mitigate these vulnerabilities, numerous countermeasures have been proposed recently in the literature. However, devising an efficient defense mechanism has proven to be a difficult task, since many approaches demonstrated to be ineffective against adaptive attackers. Thus, this article aims to provide all readerships with a review of the latest research progress on Adversarial Machine Learning in Image Classification, nevertheless, with a defender’s perspective. This article introduces novel taxonomies for categorizing adversarial attacks and defenses, as well as discuss possible reasons regarding the existence of adversarial examples. In addition, relevant guidance is also provided to assist researchers when devising and evaluating defenses. Finally, based on the reviewed literature, this article suggests some promising paths for future research.

Recent Advance on Drug Therapy Related to Myocardial Ischemia Reperfusion Injury

Myocardial ischemia reperfusion injury (MIRI) means complete or partial artery obstruction of coronary artery, and ischemic myocardium will be recirculating in a period of time. Although the ischemic myocardium can be restored to normal perfusion, its tissue damage will instead be progressive. An aggravated pathological process. MIRI is a complex entity where many inflammatory mediators play different roles, both to enhance myocardial infarction-derived damage and to heal injury. Therefore, the research and development of drugs for the prevention and treatment of this period has also become the focus. This article first studied pathophysiology of MIRI, and reviewed the research progress of MIRI-related drugs. Research results show that: MIRI is inevitable for myocardial ischemia, with the possible to double damage via the ischemic condition. Therefore, it is a serious complication and one of the most popular diseases in the world. It has always been difficult to find an effective treatment for this disease, because it is difficult to explore the inflammation behind its pathophysiology.

Current Advances in Mechanisms and Treatment of Dry Eye Disease: Toward Anti-inflammatory and Immunomodulatory Therapy and Traditional Chinese Medicine

Dry eye is currently one of the most common ocular surface disease. It can lead to ocular discomfort and even cause visual impairment, which greatly affects the work and quality of life of patients. With the increasing incidence of dry eye disease (DED) in recent years, the disease is receiving more and more attention, and has become one of the hot research fields in ophthalmology research. Recently, with the in-depth research on the etiology, pathogenesis and treatment of DED, it has been shown that defects in immune regulation is one of the main pathological mechanisms of DED. Since the non-specific and specific immune response of the ocular surface are jointly regulated, a variety of immune cells and inflammatory factors are involved in the development of DED. The conventional treatment of DED is the application of artificial tears for lubricating the ocular surface. However, for moderate-to-severe DED, treatment with anti-inflammatory drugs is necessary. In this review, the immunomodulatory mechanisms of DED and the latest research progress of its related treatments including Chinese medicine will be discussed.

International Research Progress and Evolution Trend of Interpersonal Trust—Prospects under COVID-19 Pandemic

Retrospecting articles on interpersonal trust is of great importance for understanding its current status and future development in the context of the COVID-19 pandemic, especially, with the widespread use of Big Data and Blockchain. In total, 1532 articles related to interpersonal trust were collected as research database to draw keyword co-occurrence mapping and timeline mapping by VOSviewer and CiteSpace. On this basis, the research content and evolution trend of interpersonal trust were systematically analyzed. The results show that: (1) Data cleaning by code was first integrated with Knowledge Mapping and then used to review the research of interpersonal trust; (2) Developed countries have contributed the most to the research of interpersonal trust; (3) Social capital, knowledge sharing, job and organizational performance, Chinese Guanxi are the research hotspots of interpersonal trust; (4) The research hotspots on interpersonal trust evolve from the level of individual psychology and behavior to the level of social stability and development and then to the level of organization operation and management; (5) At present, the research on interpersonal trust is in the outbreak period; fMRI technology and Big Data and Blockchain technology gradually become vital research tools of interpersonal trust, which provides significant prospects for the following research of interpersonal trust under the COVID-19 pandemic.