Search Results for author: Anna Dawid

Found 6 papers, 4 papers with code

Deconstructing the Goldilocks Zone of Neural Network Initialization

no code implementations5 Feb 2024 Artem Vysogorets, Anna Dawid, Julia Kempe

The second-order properties of the training loss have a massive impact on the optimization dynamics of deep learning models.

Unveiling the Hessian's Connection to the Decision Boundary

1 code implementation12 Jun 2023 Mahalakshmi Sabanayagam, Freya Behrens, Urte Adomaityte, Anna Dawid

Based on this finding, we provide a new and straightforward approach to studying the complexity of a high-dimensional decision boundary; show that this connection naturally inspires a new generalization measure; and finally, we develop a novel margin estimation technique which, in combination with the generalization measure, precisely identifies minima with simple wide-margin boundaries.

Introduction to Latent Variable Energy-Based Models: A Path Towards Autonomous Machine Intelligence

no code implementations5 Jun 2023 Anna Dawid, Yann Lecun

Current automated systems have crucial limitations that need to be addressed before artificial intelligence can reach human-like levels and bring new technological revolutions.

Self-Driving Cars

Unsupervised machine learning of topological phase transitions from experimental data

1 code implementation14 Jan 2021 Niklas Käming, Anna Dawid, Korbinian Kottmann, Maciej Lewenstein, Klaus Sengstock, Alexandre Dauphin, Christof Weitenberg

Recently, machine learning methods have been shown to be an alternative way of localising phase boundaries also from noisy and imperfect data and without the knowledge of the order parameter.

Anomaly Detection Quantum Gases Disordered Systems and Neural Networks Mesoscale and Nanoscale Physics Quantum Physics

Magnetic properties and quench dynamics of two interacting ultracold molecules

1 code implementation22 Oct 2020 Anna Dawid, Michał Tomza

We theoretically investigate the magnetic properties and nonequilibrium dynamics of two interacting ultracold polar and paramagnetic molecules in a one-dimensional harmonic trap in external electric and magnetic fields.

Atomic Physics Quantum Gases Chemical Physics Quantum Physics

Phase Detection with Neural Networks: Interpreting the Black Box

1 code implementation9 Apr 2020 Anna Dawid, Patrick Huembeli, Michał Tomza, Maciej Lewenstein, Alexandre Dauphin

Neural networks (NNs) normally do not allow any insight into the reasoning behind their predictions.

Quantum Physics Disordered Systems and Neural Networks

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