Elise Özalp

Elise Özalp

she/her

PhD Student at Imperial College London Amelia Earhart Fellow

Publications

You can also find my articles on my Google Scholar profile.

Publications

Inferring stability properties of chaotic systems on autoencoders’ latent spaces

Machine Learning and the Physical Sciences Workshop, NeurIPS, 2024

This work demonstrates the CAE-ESN model’s ability to infer stability properties of chaotic systems in a low-dimensional latent space, using Lyapunov exponents and covariant Lyapunov vectors to represent the geometry of the tangent space.

Citation: Özalp, E., & Magri, L. (2024). Inferring stability properties of chaotic systems on autoencoders’ latent spaces. Machine Learning and the Physical Sciences Workshop, NeurIPS 2024.
Download Paper

Stability analysis of chaotic systems in latent spaces

arXiv preprint arXiv:2410.00480, 2024

This study demonstrates a latent-space approach using a CAE-ESN model to analyze the stability properties of chaotic partial differential equations, with a focus on the Kuramoto-Sivashinsky equation.

Citation: Özalp, E., & Magri, L. (2024). Stability analysis of chaotic systems in latent spaces. In Review. arXiv preprint arXiv:2410.00480.
Download Paper

Reconstruction, forecasting, and stability of chaotic dynamics from partial data

Chaos: An Interdisciplinary Journal of Nonlinear Science, 2023

This paper explores LSTMs for reconstructing and forecasting from partial chaotic observations, and analyses their stability properties.

Citation: Özalp, E., Margazoglou, G., & Magri, L. (2023). Reconstruction, forecasting, and stability of chaotic dynamics from partial data. Chaos: An Interdisciplinary Journal of Nonlinear Science, 33(9).
Download Paper

Trends in Explainable Artificial Intelligence for Non-Experts

KI-Kritik/AI Critique Volume 4, 2023

This article from my M.Sc. examines the developments in eXplainable Artificial Intelligence (XAI) aimed at improving interpretability for non-expert users, with a focus on biases related to gender, race, and socioeconomic status.

Citation: Özalp, E., Hartwig, K., & Reuter, C. (2023). Trends in Explainable Artificial Intelligence for Non-Experts. KI-Kritik/AI Critique Volume 4, 223.
Download Paper

Physics-informed long short-term memory for forecasting and reconstruction of chaos

International Conference on Computational Science 2023, 2023

This paper introduces the physics-informed LSTM (PI-LSTM) to reconstruct and forecast unmeasured variables in chaotic systems, constrained by governing equations to improve stability and predictive accuracy.

Citation: Özalp, E., Margazoglou, G., & Magri, L. (2023). Physics-informed long short-term memory for forecasting and reconstruction of chaos. In International Conference on Computational Science (Vol. 10476, pp. 382–389). Springer, Cham.
Download Paper