International Journal of Scientific Research and Engineering Development

Blockchain systems like Bitcoin rely on Proof-of-Work (PoW), where large amounts of computational power are used to solve complex mathematical puzzles. While this approach is effective for maintaining security, it consumes significant energy and produces no meaningful output. This paper explores an alternative approach called Proof-of-Learning (PoL), where computational effort is redirected toward training machine learning models. In this framework, multiple participants collaboratively train a shared model using federated learning, allowing data to remain private on individual devices. Instead of simply demonstrating computational effort, each participant contributes by improving the model. A validation mechanism is introduced to ensure that these contributions are genuinely useful before being incorporated into the global model. The goal is to transform traditionally wasted computational resources into productive work while preserving the security and fairness of decentralized systems. By combining blockchain with artificial intelligence, this approach aims to make distributed computing more efficient, practical, and meaningful.

Aditya Kandari, Sumanth Bhargav Reddy, Sai Charan Kamarthi, Chetan Meluseema, Anita Patil,"Proof-of-Learning: A Federated Approach for Decentralized AI Model Training" International Journal of Scientific Research and Engineering Development, V9(3): Page(160-167) May-June 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.