ML-based Text Steganography
Event details
Date | 09.07.2019 |
Hour | 11:30 › 13:30 |
Speaker | Andreas Hug |
Location | |
Category | Conferences - Seminars |
EDIC candidacy exam
Exam president: Prof. Robert West
Thesis advisor: Prof. Katerina Argyraki
Thesis co-advisor: Prof. Martin Jaggi
Co-examiner: Prof. Carmela Troncoso
Abstract
Linguistic steganography systems are becoming more important with rising concerns and awareness
for user privacy. The field of natural language processing has recently witnessed tremendous
improvements in a multitude of applications that could potentially be useful.
In this report, we discuss an existing linguistic steganography approach, a new encoder-decoder
architecture and an instance of a generative adversarial network for discrete sequences. We discuss
how these three methods can be combined to create better linguistic stegosystems. Lastly, we give
a glimpse on the work done by us during this academic year as well as a short outlook on future
work.
Background papers
Matryoshka: Hiding Secret Communication in Plain Sight, by Safaka, I., Fragouli, C., Argyraki K.
Attention is all you need, by Vaswani, A., et al.
SeqGAN: Sequence Generative Adversarial Nets with Policy, by Yu, L., et al.
Exam president: Prof. Robert West
Thesis advisor: Prof. Katerina Argyraki
Thesis co-advisor: Prof. Martin Jaggi
Co-examiner: Prof. Carmela Troncoso
Abstract
Linguistic steganography systems are becoming more important with rising concerns and awareness
for user privacy. The field of natural language processing has recently witnessed tremendous
improvements in a multitude of applications that could potentially be useful.
In this report, we discuss an existing linguistic steganography approach, a new encoder-decoder
architecture and an instance of a generative adversarial network for discrete sequences. We discuss
how these three methods can be combined to create better linguistic stegosystems. Lastly, we give
a glimpse on the work done by us during this academic year as well as a short outlook on future
work.
Background papers
Matryoshka: Hiding Secret Communication in Plain Sight, by Safaka, I., Fragouli, C., Argyraki K.
Attention is all you need, by Vaswani, A., et al.
SeqGAN: Sequence Generative Adversarial Nets with Policy, by Yu, L., et al.
Practical information
- General public
- Free