secretflow/secretflow

Features

• Multi-Party Computation Frameworks - Provides a comprehensive framework for executing secure multi-party computation across distributed nodes.
• Federated Orchestration - Coordinates training tasks and communication for federated learning on partitioned data.
• Privacy-Preserving Model Training - Implements training methodologies that keep raw data on local devices while sharing only model updates to maintain privacy.
• Distributed Model Development - Enables the development of models across partitioned datasets held by different parties without revealing raw information.
• Distributed Privacy Queries - Performs analytical queries across multiple distrusting parties without exposing underlying raw data.
• Secure Distributed Analytics - Executes analytical queries and data processing across clusters to keep sensitive information hidden from the processor.
• Joint Data Analysis - Executes shared analytical workloads using cryptographic protocols across collaborative data resources.
• Privacy-Preserving Analysis Tools - Performs analytical queries and data processing across distrusting parties using secure computation.
• Encrypted Data Processing - Provides capabilities to run analysis algorithms directly on encrypted or partitioned data to keep raw information hidden from the processor.
• Multi-Party Computation - Enables joint data analysis across multiple distrusting parties without exposing underlying raw information.
• Privacy Computing Frameworks - Provides a comprehensive framework for secure multi-party computation and privacy-preserving machine learning across distributed nodes.
• Privacy-Preserving Machine Learning - Trains machine learning models on distributed datasets while keeping raw source data encrypted.
• Model Partitioning - Splits machine learning model weights and data across parties to ensure no single entity sees the whole dataset.
• Secure Multi-Party Computation Platforms - Provides a platform using cryptographic protocols to analyze joint datasets without exposing raw information.
• Decentralized Cluster Coordination - Provides peer-to-peer coordination of compute tasks across multiple parties without a central authority.
• Privacy Orchestration - Provides coordination frameworks to manage privacy-preserving computation workloads across a cluster.
• Distributed Task Schedulers - Implements distributed task scheduling based on data locality and security constraints.
• Computation Orchestration - Coordinates the sequence of distributed computation steps across isolated environments to maintain data sovereignty.
• Private Workflow Management - Manages data processing and model training as a unified sequence of operations across secure devices.
• Distributed Cluster Coordination - Coordinates secure workloads and cryptographic tasks across a cluster of independent nodes.
• Task Coordinations - Synchronizes secure computation workflows and tracks task completion across distributed nodes.

Star history