Search Results for author: Nathan Buskulic

Found 4 papers, 1 papers with code

Recovery Guarantees of Unsupervised Neural Networks for Inverse Problems trained with Gradient Descent

no code implementations8 Mar 2024 Nathan Buskulic, Jalal Fadili, Yvain Quéau

Advanced machine learning methods, and more prominently neural networks, have become standard to solve inverse problems over the last years.

Convergence and Recovery Guarantees of Unsupervised Neural Networks for Inverse Problems

no code implementations21 Sep 2023 Nathan Buskulic, Jalal Fadili, Yvain Quéau

Neural networks have become a prominent approach to solve inverse problems in recent years.

Convergence Guarantees of Overparametrized Wide Deep Inverse Prior

no code implementations20 Mar 2023 Nathan Buskulic, Yvain Quéau, Jalal Fadili

Neural networks have become a prominent approach to solve inverse problems in recent years.

Maximizing Drift is Not Optimal for Solving OneMax

1 code implementation16 Apr 2019 Nathan Buskulic, Carola Doerr

More precisely, we show that for most fitness levels between $n/2$ and $2n/3$ the optimal mutation strengths are larger than the drift-maximizing ones.

Cannot find the paper you are looking for? You can Submit a new open access paper.