GlossaryTechnical Reference

Entry point for external applications to interact with ML Bridge through REST and WebSocket endpoints.

Participants who contribute computational resources to execute ML tasks, earning rewards for processing.

Agreement among multiple compute providers about the correctness of a task result.

Smart contract implementing result verification through multi-provider consensus.

Mathematical evidence verifying computation correctness without revealing sensitive information.

Consensus-Verified Computation Protocol - ML Bridge's mechanism ensuring result quality through verification.

Security achieved through financial incentives and penalties rather than trust.

Transaction costs paid in ETH to execute operations on the blockchain.

Using a trained model to make predictions or generate outputs from new input data.

InterPlanetary File System - distributed storage protocol for AI models and data.

Time between task submission and result delivery, typically measured in seconds.

Scaling solution on Base that reduces costs and increases throughput for ML Bridge operations.

Descriptive information about AI models including capabilities, requirements, and pricing.

A decentralized platform connecting AI model providers, compute resources, and task requesters through blockchain technology.

The native utility token of ML Bridge, used for payments, staking, and rewards.

The core smart contract orchestrating task submission, provider selection, and result verification.

AI researchers and companies who deploy trained machine learning models on the platform, earning rewards based on usage.

Teaching a machine learning algorithm using data to create a model capable of predictions.

Smart contract managing AI model registration, metadata, and versioning.

Security mechanism requiring multiple parties to approve transactions.

ML Bridge becomes more valuable as more participants join the network.

Computing system inspired by biological neural networks, consisting of interconnected nodes.

A compute provider's software installation that connects to the ML Bridge network.

Smart contract managing token rewards, penalties, and staking mechanisms.

Penalty mechanism where staked tokens are confiscated for malicious behavior.

Self-executing contracts with terms written into code, running on the blockchain to automate operations.

Locking MLB tokens to participate as a provider, enabling penalty enforcement for bad behavior.

A machine learning job submitted to the network, containing input data, model selection, and payment.

Individuals or organizations who submit machine learning tasks to the ML Bridge network, paying for computation.

Smart contract handling task creation, assignment, and lifecycle management.

Number of tasks the network can process per unit time, measured in tasks per hour.

Network participants who validate AI task results through consensus mechanisms, ensuring quality and accuracy.

Earning additional MLB tokens by providing liquidity or participating in platform activities.