Li et al. proposed Proof of Training (PoT) in their paper, repurposing PoW mining power for verifiable ML training while preserving PoW incentives and security. Experiments show high throughput, strong robustness, and improved network security. Li等人在论文中提出了“训练证明”（PoT）协议，将PoW挖矿算力重用于可验证的机器学习模型训练，同时保留PoW的激励与安全性。实验表明，该方案在任务吞吐量、鲁棒性和网络安全方面具有显著潜力。 Li らは論文で、PoWマイニングの計算能力を検証可能な機械学習トレーニングに転用する「Proof of Training（PoT）」プロトコルを提案した。実験では、高いタスクスループット、堅牢性、ネットワークセキュリティの向上が示された。
Notes
PoT repurposes PoW power for verifiable ML training while preserving mining incentives.
Theoretically identifies optimal blockchain structure for training reliability.
Implements a decentralized training network prototype with high throughput.
System shows strong robustness against malicious node attacks.
PoT improves energy efficiency of PoW networks.
Privacy-efficiency trade-offs in training verification remain to be addressed.