Yesterday I delivered a talk at Kellogg College, Oxford on "Optimising Sustainable Energy with Functional Programming". The slides are now available here:
https://www.cse.chalmers.se/~patrikj/talks/Jansson_OptiFun_Kellogg_2024.pdf
and a recording here
https://www.youtube.com/watch?v=ZIP0FMDSDOA&list=PLf5C73P7ab-4OCYhM1STDCgki_pUTD7w8&index=9
It was basically a popular science talk / lecture on optimisation of (numerical) simulations. I presented two motivating example domains: fusion energy and climate impact research and went on to explain some of the concepts and techniques. Part one was about cost functions and Bayesian optimisation (balancing exploration and exploitation). Part two was about true multi-objective optimisation where the Pareto-front captures trade-offs between conflicting objectives. Finally I presented a bit about ongoing and future research directions: Pareto front computation using adaptive mesh refinement, and applications to optimal climate decisions.
Blog post:
https://patrikja.owlstown.net/posts/3188-talk-optimising-sustainable-energy-with-functional-programming
=> View attached media | View attached media
=> More informations about this toot | More toots from patrikja@functional.cafe
The video recording is now available on YouTube:
https://www.youtube.com/watch?v=ZIP0FMDSDOA&list=PLf5C73P7ab-4OCYhM1STDCgki_pUTD7w8&index=9
=> More informations about this toot | More toots from patrikja@functional.cafe This content has been proxied by September (3851b).Proxy Information
text/gemini