Hausdorff dimension, heavy tails, and generalization in neural networks*
From MaRDI portal
Publication:5020059
DOI10.1088/1742-5468/ac3ae7OpenAlexW3102590949MaRDI QIDQ5020059
Ozan Sener, George Deligiannidis, Murat A. Erdogdu, Umut Şimşekli
Publication date: 3 January 2022
Published in: Journal of Statistical Mechanics: Theory and Experiment (Search for Journal in Brave)
Full work available at URL: https://arxiv.org/abs/2006.09313
Cites Work
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Feller processes generated by pseudo-differential operators: On the Hausdorff dimension of their sample paths
- Multifractality of jump diffusion processes
- The Hausdorff dimension of the range of the Lévy multistable processes
- Density and dimension
- Learning figures with the Hausdorff metric by fractals -- towards computable binary classification
- Ergodicity of stochastic differential equations with jumps and singular coefficients
- Bridging the gap between constant step size stochastic gradient descent and Markov chains
- Multifractal properties of sample paths of ground state-transformed jump processes
- On estimating the tail index and the spectral measure of multivariate \(\alpha\)-stable distributions
- Dimension results for sample paths of operator stable Lévy processes
- First exit times of SDEs driven by stable Lévy processes
- A dimension theorem for sample functions of processes with stable components
- Lévy matters III. Lévy-type processes: construction, approximation and sample path properties
- Fractals in Probability and Analysis
- On the Macroscopic Fractal Geometry of Some Random Sets
- Flat Minima
- Some Theorems on Stable Processes
- On the ψ-Mixing Condition for Stationary Random Sequences
- 10.1162/153244302760200704
- How Much Does Your Data Exploration Overfit? Controlling Bias via Information Usage
- Cooling down Lévy flights
- Understanding Machine Learning
- Some methods of speeding up the convergence of iteration methods
- Theoretical Guarantees for Approximate Sampling from Smooth and Log-Concave Densities