The startup develops data insight software that utilizes machine learning to monitor and detect real-time data extraction, allowing organizations to maintain data privacy. By enabling knowledge sharing without compromising sensitive information, the software helps clients protect their data assets while still gaining valuable insights.
Funding
$500K raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Product
Problem
Many organizations are unable to fully leverage machine learning due to data privacy regulations, security concerns, and the inability to combine datasets from disparate sources. Traditional centralized machine learning techniques require uploading all local datasets to a single server, creating privacy and security risks.
Solution
Acuratio offers a multicloud federated learning platform that enables organizations to unlock the value of data by combining datasets without compromising privacy. The platform supports horizontal learning, vertical learning, and private analytics, allowing for the development of more accurate models and the computation of aggregate statistics while preserving both model and data privacy. It facilitates training machine learning models across multiple edge devices or servers holding local data samples, without exchanging the data itself. The platform's ease of use allows users to connect data, train models, and deploy them into production with just a few lines of code.
Target Audience
The primary target audience includes organizations in industries such as finance, healthcare, and research that require secure and private machine learning capabilities across multiple data sources and locations.
Features
- Supports horizontal federated learning with Federated Averaging and Split Learning for TensorFlow and Keras.
- Enables vertical federated learning to combine data sources with a common set of users but different features using Split Learning and Private Set Intersection.
- Facilitates private analytics with Private Join and Compute to compute aggregate statistics or segment audiences while keeping individual information private.
- Offers differential privacy integration during training.
- Includes a secure aggregation protocol for cryptographically private federated learning.
- Provides role-based authentication for granular control over access to data, models, and results.
- Supports focused updates with sparse ternary compression and random rotation matrices.
- Offers GPU support for training and serving performance and efficiency.
- Enables collaboration with role-based access control and team management.
- Provides audit logs to monitor access to data.
- Supports multicloud, on-premise, and hybrid deployments.
- Allows seamless deployment of ML models.