Publications

Microcanonical Langevin Ensembles: Advancing the Sampling of Bayesian Neural Networks

Paths and Ambient Spaces in Neural Loss Landscapes

  • Daniel Dold, Julius Kobialka, Nicolai Palm, Emanuel Sommer, David Rügamer, Oliver Dürr
  • Paper to be published @ AISTATS 2025
  • Approximate Bayesian Inference | Sampling based Inference | Loss Paths

Connecting the Dots: Is Mode-Connectedness the Key to Feasible Sample-Based Inference in Bayesian Neural Networks?

  • Emanuel Sommer*, Lisa Wimmer*, Theodore Papamarkou, Ludwig Bothmann, Bernd Bischl, David Rügamer
  • Paper @ ICML 2024
  • Bayesian Neural Networks | Sampling based Inference | MCMC | Uncertainty Quantification

Vine Copula based Portfolio Level Conditional Risk Measure Forecasting

* Equal contribution

Software

{portvine}

Package for the portfolio level unconditional as well as conditional risk measure estimation for backtesting and stress testing using Vine Copula and ARMA-GARCH models.

CRAN | GitHub | Documentation