Home — Research — Teaching — Academic services — Projects — Education — Work Experience

About me

Hi there!

I am a researcher in Machine Learning and Artificial Intelligence. Currently, I work as an Applied Scientist at Amazon. Before that, I was a postdoctoral researcher at the University of Luxembourg where I was a member of SECAN-Lab team.

Feel free to browse this website to learn about my research, teaching, and other experiences. You can also find out more by checking my LinkedIn and Google Scholar pages.

Contact:

marharyta dot aleksandrova at gmail dot com

News:

2022.11.17 Our paper k-Pareto Optimality-Based Sorting with Maximization of Choice and Its Application to Genetic Optimization by Jean Ruppert, Marharyta Aleksandrova and Thomas Engel was published in MDPI open access journal Algorithms. Resources: paper, related code, certificate

2022.06.30 Tutorial at PFIA-2022. Resources: code

2022.01.18 Our paper k-Pareto Optimality-Based Sorting with Maximization of Choice by Jean Ruppert, Marharyta Aleksandrova and Thomas Engel was accepted for The 25th International Conference on Artificial Intelligence and Statistics AISTATS-2022. Resources: preprint, code, video presentation, poster

2021.11.01 On November 3, 10 and 24 I will be presenting the theory of Causal Inference & Causal Learning at Machine Learning Seminar organized by Legato team from the University of Luxembourg. Resources: Recording of Lecture 1, Recording of Lecture 2, Recording of Lecture 3

2021.10.11 Our paper The Effect of Noise Level on the Accuracy of Causal Discovery Methods with Additive Noise Models by Benjamin Kap, Marharyta Aleksandrova and Thomas Engel was accepted for Joint International Scientific Conferences on AI BNAIC/BENELEARN 2021. Resources: presentation

2021.09.22: Check this google colab notebook (or this verion on git) to see how to develop conformal learning theory from scratch. Presented at Machine Learning Seminar organized by Legato team from the University of Luxembourg.

2021.09.11: Our paper Causal Identification with Additive Noise Models: Quantifying the Effect of Noise by Benjamin Kap, Marharyta Aleksandrova and Thomas Engel was accepted for 10èmes Journées Francophones sur les Réseaux Bayésiens et les Modèles Graphiques Probabilistes JFRB 2021. Resources: paper, presentation

2021.07.26: Our paper Impact of Model-Agnostic Nonconformity Functions on Efficiency of Conformal Classifiers: an Extensive Study by Marharyta Aleksandrova and Oleg Chertov was accepted for 10th Symposium on Conformal and Probabilistic Prediction with Applications COPA 2021. Resources: paper

2021.07.07: Our paper How Nonconformity Functions and Difficulty of Datasets Impact the Efficiency of Conformal Classifiers by Marharyta Aleksandrova and Oleg Chertov was accepted for a poster presentation at the 2021 Distribution-Free UQ workshop at ICML 2021. Resources: video presentation, paper, poster

2021.06: In July 2021 I will be giving a course “Recommender Systems: the Fundamentals” at Lviv Data Science Summer School.

2021.06: On June 29 I will be giving a tutorial “Vers l’apprentissage automatique causal” / “Towards Causal Machine Learning” at PFIA’21.

2021.05: Our paper k-Pareto Optimality for Many-Objective Genetic Optimization by Jean Ruppert, Marharyta Aleksandrova and Thomas Engel was accepted as a late-breaking abstract for The Genetic and Evolutionary Computation Conference GECCO-2021. Resources: poster, video presentation

2020.02: Our paper BacAnalytics: A Tool to Support Secondary School Examination in France by Azim Roussanaly, Marharyta Aleksandrova and Anne Boyer was accepted for 25th International Symposium on Methodologies for Intelligent Systems ISMIS 2020. Resources: video presentation, paper

2020.01: Our paper Security and Performance Implications of BGP Rerouting-resistant Guard Selection Algorithms for Tor by Asya Mitseva, Marharyta Aleksandrova, Thomas Engel and Andriy Panchenko was accepted for 35th International Conference on ICT Systems Security and Privacy Protection IFIP SEC 2020. Resources: presentation