https://p-quintanilla.github.io/post/NewsHugo Blox Builder (https://hugoblox.com)en-usFri, 01 Aug 2025 00:00:00 +0000https://p-quintanilla.github.io/media/logo_hu_5a4648074b637617.pnghttps://p-quintanilla.github.io/post/https://p-quintanilla.github.io/post/llm-symbolic-regression-2025/Thu, 09 Jul 2026 00:00:00 +0000https://p-quintanilla.github.io/post/llm-symbolic-regression-2025/<p>Can Large Language Models accelerate scientific model discovery? 🤖</p> <p>This is one of the questions we&rsquo;re exploring in our latest work on LLM-guided symbolic regression for kinetic model discovery, with Roberto Aliaga Medina and Ehecatl Antonio del Rio Chanona.</p> <p>The idea is to combine the strengths of both: symbolic regression recovers interpretable equations from data, while the LLM brings in scientific knowledge to steer the search toward models that actually make physical sense.</p> <p>Across four case studies, our framework:</p> <ul> <li>cut the number of new experiments needed to find the ground-truth model by 42–79% vs. standard symbolic regression</li> <li>had the LLM directly propose the correct model structure in half of the guided runs</li> <li>matched baseline predictive accuracy → so fewer experiments, no loss in quality</li> </ul> <p>I presented this (ongoing) work at the 9th Machine Learning and AI in (Bio)Chemical Engineering Conference in Cambridge — thanks so much to the organisers for such a great event! 😊</p> <p>Huge kudos to our brilliant Roberto Aliaga Medina, who has done all of this while still finishing his MSc in Chile and working with us in parallel!</p> <p>🔗 <a href="https://www.linkedin.com/posts/paulinaquintanilla_aiforscience-largelanguagemodels-scientificmachinelearning-ugcPost-7480957345636909058-PHJ7/?utm_source=share&amp;utm_medium=member_desktop&amp;rcm=ACoAABVxaAABUyjSi_mbKwqgIF2F8ffBOna0wIA" target="_blank" rel="noopener">View full post on LinkedIn</a></p>https://p-quintanilla.github.io/post/msca-fellowship-2026/Mon, 01 Jun 2026 00:00:00 +0000https://p-quintanilla.github.io/post/msca-fellowship-2026/<p>We are looking for outstanding postdoctoral researchers interested in applying for the <strong>MSCA Postdoctoral Fellowship</strong> (deadline: <strong>9 September 2026</strong>) to join us at the UCL Department of Chemical Engineering.</p> <p>If your research interests align with our areas — machine learning, optimisation, control, or AI for science — and you are considering applying, please get in touch to discuss potential project ideas.</p> <p>📧 <a href="mailto:p.quintanilla@ucl.ac.uk">p.quintanilla@ucl.ac.uk</a></p>https://p-quintanilla.github.io/post/ifac-world-congress-2026/Fri, 01 May 2026 00:00:00 +0000https://p-quintanilla.github.io/post/ifac-world-congress-2026/<p>Our paper <em>Hierarchical control via MPC-RL for multi-timescale battery systems</em> by Rasa Pourjam, Antonio Del Rio Chanona and Paulina Quintanilla has been accepted at the <strong>23rd IFAC World Congress</strong>.</p>https://p-quintanilla.github.io/post/loops-website-launch/Fri, 01 Aug 2025 00:00:00 +0000https://p-quintanilla.github.io/post/loops-website-launch/<p>Welcome to the LOOPS group website! We are the <strong>Learning and Optimisation Of Process Systems</strong> research group, based in the Department of Chemical Engineering at University College London.</p> <p>Stay tuned for news on publications, team updates, and opportunities to join us.</p>