You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
LTNjax is a neurosymbolic framework that allows the implementation of knowledge in the form of logical expressions as objective for neural networks. LTN uses a differentiable first-order logic language, called Real Logic, to incorporate machine learning and logic.
This software extends the Python markdown implementation with multiline table support. It can be easily used with Python-Markdown and MkDocs as well as Material for MkDocs.
SCAN (Software for Complex system Analysis via Networks): A python package implementing the machine learning technique Reservoir Computing and related methods.
Code for the paper "Franke et al.: Revisiting Neural Activation Coverage for Uncertainty Estimation", accepted for a poster session @ ESANN 2026. Contains minimal torch reimplementation of https://github.com/BierOne/ood_coverage, extended by a novel formulation for regression problems. Only the uncertainty estimation function is re-implemented.