Semiconductor USA
Back to Careers

Intern 2020-2021 - Privacy-Preserving Deep Learning

Location: Paris, France
Job Category: Internship

Apply Now
2020-2021 - Privacy-Preserving Deep Learning

AI/Machine Learning Internship - Samsung Strategy and Innovation Center - Paris - 6 months
The internship is 6 months base in the Paris office of Samsung Strategy and Innovation Center (SSIC).
----------------------------------------------
Privacy-Preserving Deep Learning
----------------------------------------------
Machine learning, especially its subfield of Deep Learning, had many amazing advances in the recent years. However, machine learning
algorithms require large amounts of data before they begin to give useful results. Data are currently mostly aggregated in large non-encrypted,
private, and centralized storage. This situation raises important privacy issues. The producer of the data has very few access to its own data (to
modify it, or remove it), and the general access to such private and sensible data is still difficult to control. Such privacy conflicts may slow down
the adoption of neural networks in sensitive domains such as healthcare or customized models on user specific data. To overcome this issue,
several privacy-preserving technologies have emerged such as Federated Learning, Secure Multiparty Computation, and Differential Privacy. A
paradigm where the user keeps the ownership and privacy of its data, but where the data of multiple users can still be used to build a general
model owned by a third party, is possible and a solution to foster machine learning algorithms adoption.
The internship goal is to work on this topic and demonstrate to help create a safer and private paradigm to allow private data owner to leverage
the recent machine learning advances. After a first step to study state of the art, the intern will leverage some recent tool in the machine learning
eco-system to code a first minimal viable product.
References:
https://arxiv.org/pdf/1811.04017.pdf: A generic framework for privacy preserving deep learning
https://arxiv.org/pdf/1911.12322.pdf: Crypto-Oriented Neural Architecture Design
https://arxiv.org/pdf/1912.04977.pdf: Advances and Open Problems in Federated Learning (Mostly Part 4)
Tools:
https://github.com/OpenMined/PySyft: PySyft
https://github.com/facebookresearch/CrypTen: CrypTen
https://blog.openmined.org/encrypted-training-on-mnist/: Encrypted mnist
https://mortendahl.github.io/2018/10/19/experimenting-with-tf-encrypted/: TF Encrypted
https://mortendahl.github.io/2018/10/19/experimenting-with-tf-encrypted/: Experimenting with TF Encrypted
https://www.tensorflow.org/federated: TF Federated
https://ai.googleblog.com/2017/04/federated-learning-collaborative.html: TF Federated
-------------------------------------------------------
Samsung Strategy and Innovation Center
-------------------------------------------------------
With offices in San Jose (US), Menlo Park (US), New York (US), Paris (France), Tel Aviv (Israel) and Seoul (Korea), the goal of Samsung
Strategy and Innovation Center (SSIC) is to smartly add artificial intelligence into Samsung products and to promote innovation. Our first lines of
work are the Automated Mobility and the Internet of Things, in order to seek and develop high impact solutions to revolutionize uses. We are
customer-centric, making our technologies respecting privacy. In collaboration with Samsung's business teams, SSIC brings the latest research
innovations to create products optimized by AI, and quickly accessible to users.
https://www.samsung.com/us/ssic/
Apply Now