RosenowGroup

Popular repositories Loading

1. Hessian-eigenvectors-PCA-DNN-weights Hessian-eigenvectors-PCA-DNN-weights Public

   The code used to train and analyze networks for RMT and PCA in tensorflow

   Jupyter Notebook

2. BBP-Phase-Transition-for-an-Extensive-Number-of-Outliers BBP-Phase-Transition-for-an-Extensive-Number-of-Outliers Public

   Python code for the results and figures used in the manuscript.

   Jupyter Notebook

3. approximate-symmetries-hessian approximate-symmetries-hessian Public

   Code accompanying the manuscript "Explaining Near-Zero Hessian Eigenvalues Through Approximate Symmetries in Neural Networks" (M. Kühn and B. Rosenow).

   Jupyter Notebook

4. fixed-tree-grammar-temperature fixed-tree-grammar-temperature Public

   Code accompanying the manuscript "Power-Law Correlations Without Apparent Temperature-Tuned Criticality in Fixed-Tree Hierarchical Languages" (M. Kühn and B. Rosenow).

   Python

5. classification-one-third-scaling classification-one-third-scaling Public

   Code accompanying the manuscript "A Boundary-Layer Mechanism for One-Third Scaling in Online Softmax Classification" (M. Kühn, Y. Thelge, and B. Rosenow).

   Python