Anne Draelos
Assistant Prof in BME & Computational Medicine at UMich
Real-time machine learning for adaptive neuroscience experiments
- Our new paper in @natcomms.nature.com introduces improv, a flexible software platform that integrates models with experiments in real-time. Traditional experiments collect all data first, then analyze it later. With improv, models analyze data as it streams in and actively guide what to do next.
- Free link: www.nature.com/articles/s41...
- We demonstrate this across multiple scenarios: using Bayesian optimization to map neural responses, predicting future population dynamics, and implementing closed-loop optogenetic photostimulation to target neurons based on their response properties discovered during an experiment.
-
View full threadThis was very fun work from my postdoc with Eva Naumann and @jmxpearson.bsky.social. Code: github.com/project-improv/improv.git. We update regularly as we implement new kinds of adaptive experiments with collaborators. Please reach out if you run into any issues or want to suggest new features!
