In a global climate modeling project, universities could use hierarchical federated learning coordinated through Substrate's off-chain workers, with regional aggregators handling local updates. [[ZKP/ZKP Base Layer/ZKP Blockchain/Technical Build Application Layer/ZK Circuit Workflow in Privacy-Preserving Computations|ZKPs verify update integrity]] through Substrate's verification infrastructure, [[ZKP/Research/Advanced Federated Learning/Secure Aggregation with SMPC|SMPC]] ensures secure aggregation, and [[Differential Privacy in Federated Learning|differential privacy]] protects sensitive research data. For instance, universities in different regions could contribute updates to a shared climate prediction model, with regional aggregators reducing communication overhead, [[ZKP/ZKP Base Layer/Core Concepts/Zero-Knowledge Proofs|ZKPs]] ensuring the updates are valid through both [[Ethereum Virtual Machine (EVM)]] pallet and native [[Substrate]] verification pallets, SMPC protecting research data, and differential privacy preventing inference of proprietary datasets, enabling secure, collaborative climate research leveraging Substrate's networking capabilities. See also: [[Introducing Privacy Pools for Scalability]]