Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.21118 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908676702339072 |
|---|---|
| author | Onobhayedo, Pius Oamen, Paul Osemudiame |
| author_facet | Onobhayedo, Pius Oamen, Paul Osemudiame |
| contents | Artificial intelligence is retracing the Internet's path from centralized provision to distributed creation. Initially, resource-intensive computation concentrates within institutions capable of training and serving large models.Eventually, as federated learning matures, billions of edge devices holding sensitive data will be able to collectively improve models without surrendering raw information, enabling both contribution and consumption at scale. This democratic vision remains unrealized due to certain compositional gaps; aggregators handle updates without accountability, economic mechanisms are lacking and even when present remain vulnerable to gaming, coordination serializes state modifications limiting scalability, and governance permits retroactive manipulation. This work addresses these gaps by leveraging cryptographic receipts to prove aggregation correctness, geometric novelty measurement to prevent incentive gaming, parallel object ownership to achieve linear scalability, and time-locked policies to check retroactive manipulation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_21118 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Trustless Federated Learning at Edge-Scale: A Compositional Architecture for Decentralized, Verifiable, and Incentive-Aligned Coordination Onobhayedo, Pius Oamen, Paul Osemudiame Machine Learning Artificial intelligence is retracing the Internet's path from centralized provision to distributed creation. Initially, resource-intensive computation concentrates within institutions capable of training and serving large models.Eventually, as federated learning matures, billions of edge devices holding sensitive data will be able to collectively improve models without surrendering raw information, enabling both contribution and consumption at scale. This democratic vision remains unrealized due to certain compositional gaps; aggregators handle updates without accountability, economic mechanisms are lacking and even when present remain vulnerable to gaming, coordination serializes state modifications limiting scalability, and governance permits retroactive manipulation. This work addresses these gaps by leveraging cryptographic receipts to prove aggregation correctness, geometric novelty measurement to prevent incentive gaming, parallel object ownership to achieve linear scalability, and time-locked policies to check retroactive manipulation. |
| title | Trustless Federated Learning at Edge-Scale: A Compositional Architecture for Decentralized, Verifiable, and Incentive-Aligned Coordination |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2511.21118 |