This talk covers SBI, a method for Bayesian parameter inference using simulated data, and introduces the sbi library, an open-source tool for practitioners and researchers.
Hands-on training in Simulation-Based Inference (SBI), designed for applications in neuroscience, astrophysics, and biology. Developed in collaboration with the University of Tübingen and the TransferLab at the appliedAI Institute for Europe.
We propose a new method to perform simulation-based inference for mixed data e.g., with continuous and discrete data types, like they often occur in models of decision-making.