Picture for Moritz Helias

Moritz Helias

Applications of Statistical Field Theory in Deep Learning

Add code
Feb 25, 2025
Viaarxiv icon

From Kernels to Features: A Multi-Scale Adaptive Theory of Feature Learning

Add code
Feb 05, 2025
Viaarxiv icon

Graph Neural Networks Do Not Always Oversmooth

Add code
Jun 04, 2024
Viaarxiv icon

A theory of data variability in Neural Network Bayesian inference

Add code
Jul 31, 2023
Viaarxiv icon

Speed Limits for Deep Learning

Add code
Jul 27, 2023
Figure 1 for Speed Limits for Deep Learning
Viaarxiv icon

Optimal signal propagation in ResNets through residual scaling

Add code
May 12, 2023
Figure 1 for Optimal signal propagation in ResNets through residual scaling
Figure 2 for Optimal signal propagation in ResNets through residual scaling
Figure 3 for Optimal signal propagation in ResNets through residual scaling
Figure 4 for Optimal signal propagation in ResNets through residual scaling
Viaarxiv icon

Neuronal architecture extracts statistical temporal patterns

Add code
Jan 24, 2023
Figure 1 for Neuronal architecture extracts statistical temporal patterns
Figure 2 for Neuronal architecture extracts statistical temporal patterns
Figure 3 for Neuronal architecture extracts statistical temporal patterns
Figure 4 for Neuronal architecture extracts statistical temporal patterns
Viaarxiv icon

Origami in N dimensions: How feed-forward networks manufacture linear separability

Add code
Mar 21, 2022
Figure 1 for Origami in N dimensions: How feed-forward networks manufacture linear separability
Figure 2 for Origami in N dimensions: How feed-forward networks manufacture linear separability
Figure 3 for Origami in N dimensions: How feed-forward networks manufacture linear separability
Figure 4 for Origami in N dimensions: How feed-forward networks manufacture linear separability
Viaarxiv icon

Decomposing neural networks as mappings of correlation functions

Add code
Feb 10, 2022
Figure 1 for Decomposing neural networks as mappings of correlation functions
Figure 2 for Decomposing neural networks as mappings of correlation functions
Figure 3 for Decomposing neural networks as mappings of correlation functions
Figure 4 for Decomposing neural networks as mappings of correlation functions
Viaarxiv icon

Unified Field Theory for Deep and Recurrent Neural Networks

Add code
Jan 07, 2022
Figure 1 for Unified Field Theory for Deep and Recurrent Neural Networks
Figure 2 for Unified Field Theory for Deep and Recurrent Neural Networks
Figure 3 for Unified Field Theory for Deep and Recurrent Neural Networks
Viaarxiv icon