IEEE Access , IEEE, 2022. DOI (I.F.: 3.476)
Neural networks based cryptography has seen a significant growth since the introduction of adversarial cryptography which makes use of Generative Adversarial Networks (GANs) to build neural networks that can learn encryption. The encryption has been proven weak at first but many follow up works have shown that the neural networks can be made to learn the One Time Pad (OTP) and produce perfectly secure ciphertexts. To the best of our knowledge, existing works only considered communications between two or three parties. In this paper, we show how multiple neural networks in an adversarial setup can remotely synchronize and establish a perfectly secure communication in the presence of different attackers eavesdropping their communication. As an application, we show how to build Secret Sharing Scheme based on this perfectly secure multi-party communication. The results show that it takes around 45,000 training steps for 4 neural networks to synchronize and reach equilibria. When reaching equilibria, all the neural networks are able to communicate between each other and the attackers are not able to break the ciphertexts exchanged between them.
Ph.D Symposium of the European Conference on Service-Oriented and Cloud Computing (ESOCC) 2013, September 2013.
The advent of cloud computing has provided the opportunity to externalize the identity management processes, shaping what has been called Identity Management as a Service (IDaaS). However, as in the case of other cloud-based services, IDaaS brings with it great concerns regarding security and privacy, such as the loss of control over the outsourced data. As part of this PhD thesis, we analyze these concerns and propose BlindIdM, a model for privacy-preserving IDaaS with a focus on data privacy protection through the use of proxy re-encryption.