AI as statistical methods for imperfect theoriesDownload PDF

24 Sept 2021, 16:11 (edited 25 Sept 2021)NeurIPS-AI4Science PosterReaders: Everyone
  • Keywords: epistemology, machine learning, vibration effects, model realism, pre-paradigmatic science
  • TL;DR: Science needs statistical reasoning based more on model predictions and less brittle to modeling ingredients
  • Abstract: Science has progressed by reasoning on what models could not predict because they were missing important ingredients. And yet without correct models, standard statistical methods for scientific evidence are not sound. Here I argue that machine-learning methodology provides solutions to ground reasoning about empirically evidence more on models’ predictions, and less on their ingredients.
  • Track: Original Research Track
1 Reply