Hoil Lee

activity

20212023

most cited

α

-Stable convergence of heavy-tailed infinitely-wide neural networks

3 citations · 7 across the 3 of their papers we have counts for

collaborators

3 papers

stat.ML2023★ 1 cited

Over-parameterised Shallow Neural Networks with Asymmetrical Node Scaling: Global Convergence Guarantees and Feature Learning

Francois Caron, Fadhel Ayed, Paul Jung +3

We consider gradient-based optimisation of wide, shallow neural networks, where the output of each hidden node is scaled by a positive parameter. The scaling parameters are non-ide…

stat.ML2022★ 3 cited

Deep neural networks with dependent weights: Gaussian Process mixture limit, heavy tails, sparsity and compressibility

Hoil Lee, Fadhel Ayed, Paul Jung +3

This article studies the infinite-width limit of deep feedforward neural networks whose weights are dependent, and modelled via a mixture of Gaussian distributions. Each hidden nod…

stat.ML2021★ 3 cited

$α$ -Stable convergence of heavy-tailed infinitely-wide neural networks

Paul Jung, Hoil Lee, Jiho Lee +1

We consider infinitely-wide multi-layer perceptrons (MLPs) which are limits of standard deep feed-forward neural networks. We assume that, for each layer, the weights of an MLP are…

Over-parameterised Shallow Neural Networks with Asymmetrical Node Scaling: Global Convergence Guarantees and Feature Learning

Deep neural networks with dependent weights: Gaussian Process mixture limit, heavy tails, sparsity and compressibility

α-Stable convergence of heavy-tailed infinitely-wide neural networks

$α$ -Stable convergence of heavy-tailed infinitely-wide neural networks