📣 If you've got an objective that exhibits symmetries, you should be using invariant kernel BO 📣

🚀 More sample efficient than constrained/naive BO!

🚀 More compute efficient than data augmentation!

🧵 1/4

#NeurIPS2024 #BayesianOptimisation #ai

Showcased my work on automatic calibration with nested Bayesian optimisation 🌾
Also demoed our ADD-TREES research through the planting app — amazing engagement!

Huge thanks to @idsaiexeter.bsky.social for their funding!

#AIUK #BayesianOptimisation #AgriTech #Sustainability #NetZero #ECR

Adaptive ε‑PAL swaps uniform grids for a tree‑based discretisation, speeding discovery of ε‑accurate Pareto sets and needing fewer evaluations than other methods. Read more: https://getnews.me/adaptive-e-pal-fast-multi-objective-bayesian-optimization-without-grids/ #bayesianoptimisation #pareto

A side from these projects, I collaborate on a number of projects involving tracking, sensor and data fusion in a variety of domains.
I am interested in #machinelearning #ML, #gaussianprocesses and #bayesianoptimisation (some papers might come around in the new year).