LogoLogo
search
⌘Ctrlk
LogoLogo
  • Nesa Docs
    • Introduction to Nesa
    • 1Why Nesa Represents the Future of the AI Economy
    • 2Why Centralized AI Infrastructure Is Broken
    • 3Why Decentralized AI Has Not Worked (Yet)
    • Organization of the Documentation
  • Background and Exploratory Notes
  • Major Innovations
    • 1Private Inference for AI
    • 2Decentralized Execution for AI
      • Why Decentralized AI?
      • DAIs: Building Decentralized AI Applications on Nesa
      • Models on Nesa: Playground and Custom Uploads
      • Trust and Incentives: Miner Reputation
      • Efficient Inference: MetaInf and Execution Optimization
      • Background and Exploratory Notes
        • Blockchain-based Sequential Neural Sharding (BSNS)
        • Model Partitioning and Deep Network Sharding
        • Dynamic Sharding of Arbitrary Neural Networks
        • Cache Optimization to Enhance Efficiency
        • Enhanced MTPP Slicing of Topological Order
        • BSNS with Parameter-efficient Fine-tuning via Adapters
        • Swarm Topology
        • Free-Riding Prevention
        • Validation, Reputation, and Miner Lifecycle
    • 3AI Link: Cross-chain Access for AI
    • $NES Utility
    • Background and Exploratory Notes
  • Using Nesa
    • Getting Started
    • Via Web
    • Via SDK
    • Via IBC
  • Run a Nesa Node
    • Prerequisites
    • Installation
    • Troubleshooting
    • FAQ
  • Additional Read
  • Links
    • nesa.aiarrow-up-right
    • Nesa Playgroundarrow-up-right
    • Nesa Chain Explorerarrow-up-right
    • Nesa EVM Explorerarrow-up-right
    • Nesa Discordarrow-up-right
    • Nesa Twitterarrow-up-right
    • Terms of Servicearrow-up-right
    • Privacy Policyarrow-up-right
gitbookPowered by GitBook
block-quoteOn this pagechevron-down
  1. Major Innovationschevron-right
  2. 2Decentralized Execution for AI

Background and Exploratory Notes

Blockchain-based Sequential Neural Sharding (BSNS)chevron-rightModel Partitioning and Deep Network Shardingchevron-rightDynamic Sharding of Arbitrary Neural Networkschevron-rightCache Optimization to Enhance Efficiencychevron-rightEnhanced MTPP Slicing of Topological Orderchevron-rightBSNS with Parameter-efficient Fine-tuning via Adapterschevron-rightSwarm Topologychevron-rightFree-Riding Preventionchevron-rightValidation, Reputation, and Miner Lifecyclechevron-right
PreviousEfficient Inference: MetaInf and Execution Optimizationchevron-leftNextBlockchain-based Sequential Neural Sharding (BSNS)chevron-right