A curated list of use cases and applications of Fully Homomorphic Encryption.
|2022.11.17||Sentiment analysis on Encrypted data||Sentiment Analysis on Encrypted data with Machine Learning, Natural Language Processing and Fully Homomorphic Encryption (a Hugging Face space)||Zama (Concrete-ml)|
|2022.09||Spiral for BTC||Look up any Bitcoin (BTC) address balance, without revealing the address to anyone||Spiral|
|2022.08||Ensuring security of artificial pancreas device system using homomorphic encryption||The privacy and security of a person’s health data is a human right protected by law in many countries. However, networked information systems that store and process health data may have security vulnerabilities and are attractive to attacks aimed to gain either unauthorized access to these data or compromise it. Compromising data of patients with chronic conditions like Diabetes Mellitus has potentially life-threatening consequences (e.g., from incorrect insulin dosing due to loss of glucose measurement data integrity). Consequently, privacy-preserving computing methods are called to mitigate the risk of a data breach.||Haotian Weng, Chirath Hettiarachchi, Christopher Nolan, Hanna Suominen, Artem Lenskiy|
|2022.08.18||Encrypted Conway's Game of Life||Encrypted Conway's Game of Life in Rust with the Concrete library||Zama|
|2022.08.15||Secure human action recognition by encrypted neural network inference||Advanced computer vision technology can provide near real-time home monitoring to support “aging in place” by detecting falls and symptoms related to seizures and stroke. Affordable webcams, together with cloud computing services (to run machine learning algorithms), can potentially bring significant social benefits. However, it has not been deployed in practice because of privacy concerns. In this paper, we propose a strategy that uses homomorphic encryption to resolve this dilemma, which guarantees information confidentiality while retaining action detection. Our protocol for secure inference can distinguish falls from activities of daily living with 86.21% sensitivity and 99.14% specificity, with an average inference latency of 1.2 seconds and 2.4 seconds on real-world test datasets using small and large neural nets, respectively. We show that our method enables a 613x speedup over the latency-optimized LoLa and achieves an average of 3.1x throughput increase in secure inference compared to the throughput-optimized nGraph-HE2.||Miran Kim, Xiaoqian Jiang, Kristin Lauter, Elkhan Ismayilzada, Shayan Shams|
|2022.07.22||Morfix.io||BWeb-based UI to play around with the Microsoft Seal library||Morfix|
|2022.07.07||Titanic ML from disaster||Building a predictive model over encrypted data with Concrete-ML||Zama|
|2022.06.07||lattigo-polls||Web-application for scheduling meetings using lattigo||LDS|
|2022.05.24||Spiral wiki||Read Wikipedia privately using homomorphic encryption||Samir Menon|
|2022.10.04||crypto-geofence||Geo-fencing demo application based on Paillier scheme||Georeactor|
|2022.07.08||nGraph-HE||Deep Learning (DL) with HE through Intel’s DL graph compiler nGraph based on SEAL||Intel AI|
|2022.04.26||Rosetta||A privacy-preserving framework based on TensorFlow||LatticeX Foundation|
|2022.03.10||tf-encrypted||Bridge between TensorFlow and the Microsoft SEAL||TF Encrypted|
💙 This website is a resource provided and contributed by the FHE.org community. We welcome any contributions to this website! Read the contribution guidelines first and simply open a PR on the Github repo to add your resources.