Proof of Training (PoT): Harnessing Crypto Mining Power for Distributed AI Training

Peihao Li
7/13/2023
cs.CRcs.CEcs.DC

Official code: P-HOW/proof-of-training (3 stars, MIT, last commit 2023-07-20).

Abstract

In the midst of the emerging trend of integrating artificial intelligence (AI) with crypto mining, we identify three major challenges that create a gap between these two fields. To bridge this gap, we introduce the proof-of-training (PoT) protocol, an approach that combines the strengths of both AI and blockchain technology. The PoT protocol utilizes the practical Byzantine fault tolerance (PBFT) consensus mechanism to synchronize global states. To evaluate the performance of the protocol design, we present an implementation of a decentralized training network (DTN) that adopts the PoT protocol. Our results indicate that the protocol exhibits considerable potential in terms of task throughput, system robustness, and network security.

View on arXivView PDF

Code Implementations(1)

P-HOW/proof-of-trainingOfficial
MIT

Go implementation of the Proof-of-Training (POT) protocol

30Go, SolidityJun 5, 20233 years agoMIT

Similar Papers

SEDULity: A Proof-of-Learning Framework for Distributed and Secure Blockchains with Efficient Useful Work

84% match2025

Proof-of-Collaborative-Learning: A Multi-winner Federated Learning Consensus Algorithm

84% match1 repo12024

Substituting Proof of Work in Blockchain with Training-Verified Collaborative Model Computation

81% match2025

Proof of Reasoning for Privacy Enhanced Federated Blockchain Learning at the Edge

81% match2026

Blockchain-Enabled Federated Learning

80% match1 repo392025

Survey on Strategic Mining in Blockchain: A Reinforcement Learning Approach

80% match2025

Cite this paper

@article{li2023proof,
  title  = {Proof of Training (PoT): Harnessing Crypto Mining Power for Distributed AI Training},
  author = {Peihao Li},
  year   = {2023},
  eprint = {2307.07066},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2307.07066v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2307.07066.svg)](https://codeofpaper.com/paper/2307.07066)

Discussion