Adversarial Machine Learning in Vision, Speech, and Text
In spite of the impressive (and sometimes even
superhuman) accuracies of machine learning on various
tasks such as object recognition, speech recognition,
natural language processing and playing Go, classifiers
perform badly in the presence of small imperceptible but
adversarial (opposing) perturbations in the input
sample. In addition to being an intriguing phenomenon,
the existence of such “adversarial examples” exposes a
serious vulnerability in current machine learning (ML)
systems, and questions the future of ML and Artificial
Intelligence (AI). In the current scenario, we see a
rising level of hostile behavior in many application
domains that include: email (spamming), biometric
system, voice assistant, web search and pay-per-click
advertisements to name a few. In this special session,
we invite the contributions from researchers involved in
studying the behavior of such machine learning systems
that are susceptible to attacks which disrupt the system
it was intended to benefit, and appropriate defense
mechanisms that can be incorporated to prevent such
attacks.
Some of the seminal publications in the domain of
Adversarial Machine Learning are mentioned below:
● Goodfellow, I.J., Shlens, J., Szegedy, C.: Explaining
and harnessing adversarial examples. ArXiv Preprint
ArXiv:1412.6572 (2014)
● Moosavi-Dezfooli, S.-M., Fawzi, A., Frossard, P.:
Deepfool: a simple and accurate method to fool deep
neural networks. In: Proceedings of the IEEE Conference
on Computer Vision and Pattern Recognition, pp.
2574–2582 (2016)
● Shamsabadi, A.S., Teixeira, F.S., Abad, A., Raj, B.,
Cavallaro, A., Trancoso, I.: FoolHD: Fooling speaker
identification by highly imperceptible adversarial
disturbances. In: IEEE International Conference on
Acoustics, Speech and Signal Processing (ICASSP), pp.
6159–6163 (2021)
● Zhang, W. E., Sheng, Q. Z., Alhazmi, A., & Li, C.:
Adversarial attacks on deep-learning models in natural
language processing: A survey, ACM Transactions on
Intelligent Systems and Technology (TIST), 11(3), 1-41
(2020)
Related topics include but are not limited
to:
● Black-box
and White-box Adversarial attacks
● Physically Realizable Adversarial Attacks
● Adversarial Defense
● Robust Machine Learning Models
● Robustness evaluation
● Adversarial input detection
● Transferability of Black-box Adversarial Attacks
● Adversarial Attacks on Cyber-physical Systems
● Targeted Adversarial Attacks on Critical Security Systems
● Evaluation of Model Robustness
● Spoofing
● Data Privacy and Security
● Differential Privacy
● Poisoning Attacks
Organizers: Dr. Manjunath V. Joshi,
Dhirubhai Ambani Institute of Information and Communication
Technology, India (E-mail:
mv_joshi@daiict.ac.in)
Dr. Srimanta Mandal, Dhirubhai Ambani Institute of Information and
Communication Technology, India (E-mail:
srimanta_mandal@daiict.ac.in)
Dr. Shruti Bhilare, Dhirubhai Ambani Institute of Information and
Communication Technology, India (E-mail:
shruti_bhilare@daiict.ac.in)
Dr. Avik Hati, National Institute of Technology Tiruchirappalli,
India (E-mail: avikhati@nitt.edu
/ avikhatiece@gmail.com)
Bio: Manjunath V. Joshi received a Ph.D. degree from the Indian Institute of Technology Bombay (IIT Bombay), Mumbai, India. Currently, he is serving as a Professor at Dhirubhai Ambani Institute of Information and Communication Technology, Gandhinagar, India. He has been involved in active research in the areas of Signal and Image Processing, Cognitive Radio, Computer Vision, and Machine Learning, quantum computing and has several publications in quality journals and conferences. He has co-authored four books. So far, 10 PhD students have graduated under his supervision. Dr. Joshi was a recipient of the Outstanding Researcher Award in Engineering Section by IIT Bombay in 2005 and the Dr. Vikram Sarabhai Award for the year 2006–2007 in the field of information technology constituted by the Government of Gujarat, India. He served as a Program Co-Chair for the 3rd ACCV Workshop on E-Heritage, 2014 held at Singapore. He has also served as Visiting Professor at IIT Gandhinagar and IIIT Vadodara. He has visited Germany, Italy, France, Hong Kong, USA, Canada, South Korea, Indonesia and contributed to research in his area of expertise.
Bio: Dr. Srimanta Mandal received his Ph.D. from IIT Mandi, India in 2017. He has been a postdoctoral fellow with the Department of Electrical Engineering, IIT Madras, India, from 2017 to 2018. Since October 2018, he has been with DAIICT, Gandhinagar, where he is currently an associate professor. During his PhD, he received travel grant from IIT Mandi, for presenting work at International Conference on Image Processing 2014, Paris, France. So far, he supervised 20 master’s students in their dissertation/project work, and co-supervised 1 PhD student. He has published several articles in national/international journals and conferences. He has received the best paper award (runner up) in the Indian Conference on Computer Vision, Graphics and Image Processing 2018. He served as reviewer for various conferences and journals. He served as an executive committee member of IEEE SPS Gujarat chapter from 2019 to 2022. He is a life member of IUPRAI and ISRS. His research interests include image processing, computer vision, and machine learning.
Bio: Dr. Shruti Bhilare is an Assistant Professor in Dhirubhai Ambani Institute of Information and Communication Technology, Gandhinagar, Gujarat since July 2019. She received her Ph.D. degree in Computer Science and Engineering from Indian Institute of Technology Indore (IIT Indore), India. Her research interests include pattern recognition and image processing with focus on biometric applications and adversarial machine learning. She received travel grants from DST and CSIR for presenting her research in international conferences in the USA and Japan. She has published several papers in reputed international conferences and journals and serves as reviewer for various conferences and journals.
Bio: Avik Hati is currently an Assistant Professor at National Institute of Technology Tiruchirappalli, India. He received his B.Tech. Degree in Electronics and Communication Engineering and M.Tech. Degree in Electronics and Electrical Engineering. He received his Ph.D. degree in Electrical Engineering from the Indian Institute of Technology Bombay in 2018. He was a Postdoctoral Researcher at the Pattern Analysis and Computer Vision Department of Istituto Italiano di Tecnologia, Genova, Italy. He was an Assistant Professor at Dhirubhai Ambani Institute of Information and Communication Technology (DA-IICT), Gandhinagar from 2020 to 2022. He joined National Institute of Technology Tiruchirappalli in 2022. His research interests include image and video co-segmentation, subgraph matching, saliency detection, scene analysis, robust computer vision, adversarial machine learning.