Picture for François Ged

François Ged

Matryoshka Policy Gradient for Entropy-Regularized RL: Convergence and Global Optimality

Add code
Mar 22, 2023
Viaarxiv icon

Deep Linear Networks Dynamics: Low-Rank Biases Induced by Initialization Scale and L2 Regularization

Add code
Jun 30, 2021
Figure 1 for Deep Linear Networks Dynamics: Low-Rank Biases Induced by Initialization Scale and L2 Regularization
Figure 2 for Deep Linear Networks Dynamics: Low-Rank Biases Induced by Initialization Scale and L2 Regularization
Figure 3 for Deep Linear Networks Dynamics: Low-Rank Biases Induced by Initialization Scale and L2 Regularization
Figure 4 for Deep Linear Networks Dynamics: Low-Rank Biases Induced by Initialization Scale and L2 Regularization
Viaarxiv icon

Geometry of the Loss Landscape in Overparameterized Neural Networks: Symmetries and Invariances

Add code
May 25, 2021
Figure 1 for Geometry of the Loss Landscape in Overparameterized Neural Networks: Symmetries and Invariances
Figure 2 for Geometry of the Loss Landscape in Overparameterized Neural Networks: Symmetries and Invariances
Figure 3 for Geometry of the Loss Landscape in Overparameterized Neural Networks: Symmetries and Invariances
Figure 4 for Geometry of the Loss Landscape in Overparameterized Neural Networks: Symmetries and Invariances
Viaarxiv icon