Author: Harvey Demian Bastidas Caicedo
Degree: Master's in Engineering
Institution: Pontificia Universidad Javeriana Cali
Year: 2018
Original Spanish title: Computación Evolutiva Descentralizada de Modelo Híbrido usando Blockchain y Prueba de Trabajo de Optimización
The thesis proposes an optimization proof of work for a decentralized hybrid evolutionary-computation system. Instead of spending the computational work associated with block production on a cryptographic hash puzzle, participating nodes use that capacity to improve candidate solutions. The blockchain records optimization operations for traceability and helps synchronize optimization state across an island-model distributed evolutionary algorithm.
The implementation was evaluated with a reinforcement-learning application for foreign-exchange trading automation. The experiments examined scalability, fault tolerance and rejection of invalid optimization results.
This thesis is the research origin of DOIN, the Decentralized Optimization and Inference Network. The current implementation extends the original work and is organized into three active repositories:
- doin-core: protocol models, consensus rules, cryptographic identity and plugin interfaces.
- doin-node: unified participant runtime, networking, blockchain persistence, coordination and analytics.
- doin-plugins: reference domain plugins and adapters for external optimization systems.
The repository PDF omits the first two administrative front-matter pages from the university copy because they contained personal contact information. No research chapter, abstract, result, reference or technical appendix was removed.