Meetings
-
Approaching Deep Learning through the Spectral Dynamics of Weights
-
The importance of discretisation drift in deep learning
-
Information-theoretic generalization bounds for black-box learning algorithms
-
Stochastic Collapse: How Gradient Noise Attracts SGD Dynamics Towards Simpler Subnetworks
-
Predicting grokking long before it happens
-
The geometry of neural nets' parameter spaces under reparametrization
-
Can Neural Network Memorization Be Localized?
-
DINO v1 and v2: Self-Supervised Vision Transformers
-
SGD with Large Step Sizes Learns Sparse Features
-
Bottleneck structure in large depth networks
-
Loss Landscapes are All You Need: Neural Network Generalization Can Be Explained Without the Implicit Bias of Gradient Descent
-
Understanding edge of stability
-
Lottery ticket hypothesis and its current state
-
Intrinsic Dimension, Persistent Homology and Generalization in Neural Networks
-
A Loss Curvature Perspective on Training Instability in Deep Learning
-
When Are Solutions Connected in Deep Networks?
-
From Gradient Flow on Population Loss to Learning with Stochastic Gradient Descent
-
Towards Understanding Sharpness-Aware Minimization
-
When Do Neural Networks Outperform Kernel Methods?
-
SGD: The Role of Implicit Regularization, Batch-size and Multiple-epochs
-
Does the Data Induce Capacity Control in Deep Learning?
-
Deep Ensembles: A Loss Landscape Perspective
-
Taxonomizing local versus global structure in neural network loss landscapes
-
The Effects of Mild Over-parameterization on the Optimization Landscape of Shallow ReLU Neural Networks
-
The Geometry of Neural Network Landscapes: Symmetry-Induced Saddles & Global Minima Manifold
-
Exploring Generalization in Deep Learning
subscribe via RSS