2PM. Network

Public, Privacy, Models (2PM)

Privacy-Computing AI Model Ecosystem

Problem Statement

  • Data Privacy: Growing concerns over data privacy and security.
  • Data Traceability: Difficulty in verifying the source and integrity of data and computation process.
  • AI Accessibility: No access to AI models trained by privacy data for the public.

    • 2PM Proposition - Comprehensive Privacy Computing Solution

    Data - Computing - Verification - Monetization - Expansion

    2PM. Solution - Modular Architecture

    2PM. Solution - Fully Homomorphic Encryption

    Based on ZAMA Concrete ML, 2PM.Network uses FHE to enable large-scale Individual collaborations.

    2PM. Solution - Federated Learning

    2PM.Network provides more efficient solutions for business clients and large dataset owners.

    Integration with 0G

    Demo Video

    Website:

    Competitive Landscape

    BUIDL on 0G

    IdentityContract: Manages node identities within the 2PM network. It ensures secure and authenticated participation of nodes in the federated learning process.

    HFLContract: Handles horizontal federated learning tasks, enabling collaboration among multiple nodes without sharing raw data.

    DataHub: Manages node data, ensuring secure storage, access, and retrieval of data required for federated learning.

    HLRContract: Executes horizontal logistic regression tasks, facilitating collaborative model training across multiple nodes.

    PlonkVerifier3: Verifies zero-knowledge proofs with an input length of 3, ensuring data integrity and privacy during the learning process.

    DataRegistry: Allows users to register, verify, and index datasets, thus standardizing the fields and contents that need to be encrypted and uploaded.

    Core Team

    Contact us: hi@2pm.network

    Twitter: @2PM_Network