thinkingML
High-impact work in optimisation, adaptation, and machine intelligence.
Evolutionary Cellular Optimisation (ECO)
ECO is a generative optimisation algorithm that builds its own search space during execution. By evolving hyperparameters as dynamic cellular structures, ECO treats optimisation as a process of discovery, not selection, constructing viable configurations through generative constraint rather than sampling fixed candidates.
LoRA with LLaMA-3 on FreshQA
Lightweight adaptation of LLaMA-3 8B using LoRA, guided by ECO for dynamic hyperparameter control. Evaluated on unseen QA data with deterministic metrics under 4×H100 conditions.
Cross-Domain Optimisation Evaluation
ECO assessed across ROCT, XRAY, and IMDB to evaluate generalisation, constructive search, and resilience to noise in divergent optimisation landscapes.