NetBioMed, a satellite of the NetSci conference, in their own words brings together researchers on Machine Learning, Digital Twins and Complex Systems towards solving problems in biology and medicine.
On June the 2nd 2025 our President, Dr Marco Manca, has joined friends and colleagues from around the world to discuss the epistemology of life sciences.
In an intense half an hour talk, Dr Manca has lead the audience through a brief history of delusions in medical knowledge, to immediately focus on the processes and pitfalls of building human knowledge, the impact of assumptions (dscussing the case of ergodicity assumptions) and models on measurements (and thus, on data). He discussed the pros and cons of machine learning and the limitations of the aspiration to extract "objective knowledge" from data, or even dependable knowledge, because of the very founding principles on which ML as we have it today works.
He then moved on to reflect on the shifting regimes between "modelling" and "choosing how to act", and implications of the fat tailedness emerging from uncertainty about uncertainty.
The last part of the talk was devoted to discussing the relationship between Occam's razor and entropy and information theory, what emergence means for modelling and deciding about complex systems, and why the BioDynaMo consortium has come together to build an agent-based simulation suite geared to efficiency and flexibility (spoiler: to allow for multiple runs and challenging of models/assumptions by tail distributions of unrealistic/unobseerved results).
An intense conversation, which seemed well received and we certainly hope will continue with old and new friends in the coming future.
P.S.: we hope to convince Dr Manca to record his lecture and share it with you all here some time in the future, stay tuned ^_^
Comments
Post a Comment