Towards Adversarial Machine Learning and Defenses in Sensors Applications

Dear Colleagues,

Machine Learning has seen immense growth in recent decades across a broad spectrum of applications. The emergence and evolution of deep learning have further improved the performance of many systems. The deep learning models’ evolution has a great impact on computer vision systems, medical imaging, speech recognition, robotics, and self-driving cars. The deep learning models provide excellent performance and reliability due to its ability to generalize for many fields and to identify underlying patterns and make future predictions. The main research areas for deep learning in computer vision, due to its excellent performance, is in surveillance and biometric verification systems. The emergence of adversarial machine learning provides a new direction to the machine learning field where limited data is available. The development of the Generative Adversarial Network (GAN) provides artificial data, which is usually difficult to judge even from the human eye. The artificial data generated by GAN is used for both good and bad purposes.

The ability of deep learning models to observe and recognize patterns makes them valuable in the field of computer vision and image processing. However, deep learning models can be fooled by artificial (fake) data. It has been observed that slight modifications in the images can cause deep learning models to make wrong predictions. These wrong classifications or predictions prove expensive, especially for biometric recognition including speech, face, iris, gait recognition, and autonomous vehicles. These issues make deep learning vulnerable to adversarial attacks. Therefore, there is a strong need to design defenses against such adversarial attacks and this area needs more investigation and research.

This Special Issue aims to provide a forum for individuals from academia and industry to present their novel ideas in the field of adversarial machine learning, GANs, adversarial examples, adversarial defenses, image anonymity, and image forensics. The anticipated research should be industry and application-oriented to handle important issues. Original technical research articles including review papers are invited.

Dr. Irfan Mehmood
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Sensors is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.