Pavel Ilinov

This page reports only the academic work registered in the databases of the Transport and Mobility Laboratory, and is not necessarily a comprehensive list of the work by Pavel Ilinov.

Personal webpage

Technical reports [1]

2024

Ilinov P., Matveenko A., Senkov M., Starkov E. (2024). Optimally Biased Expertise. Report TRANSP-OR 240223


@TechReport{IlinMatvSenkStar24,
  author = {Pavel Ilinov and Andrei Matveenko and Maxim Senkov and Egor Starkov},
  title = {Optimally Biased Expertise},
  institution = {Transport and Mobility Laboratory, Ecole Polytechnique F\'ed\'erale de Lausanne},
  year = {2024},
  type = {Technical Report},
  number = {TRANSP-OR 240223},
  address = {Lausanne, Switzerland}}}

Talks [6]

2024

Ilinov P., Matveenko A., Sharma S., Tsakas E., Mark V. Sequential Search with Flexible Information March 14, 2024. MaCCI Annual 2024, Mannheim Center of Competition and Innovation, Mannheim, Switzerland.


@Misc{Ilinov_MAN2024,
  author = {Pavel Ilinov and Andrei Matveenko and Salil Sharma and Elias Tsakas and Voorneveld Mark},
  title = {Sequential Search with Flexible Information},
  howpublished = {\url{https://transp-or.epfl.ch/documents/talks/Ilinov_MAN2024.pdf}},
  year = {2024},
  note = {Presented at MaCCI Annual 2024, Mannheim Center of Competition and Innovation, Mannheim, Switzerland}}}

2023

Ilinov P., Matveenko A., Sharma S. Is it better to be first? Search with endogenous information September 21, 2023. Invited talk, IOB, University of Basel, University of Basel, Switzerland.


@Misc{Ilinov_BAS2023,
  author = {Pavel Ilinov and Andrei Matveenko and Salil Sharma},
  title = {Is it better to be first? Search with endogenous information},
  howpublished = {\url{https://transp-or.epfl.ch/documents/talks/Ilinov_BAS2023.pdf}},
  year = {2023},
  note = {Presented at Invited talk, IOB, University of Basel, University of Basel, Switzerland}}}

Ilinov P., Matveenko A., Sharma S. Is it better to be first? Search with endogenous information August 30, 2023. EEA-ESEM 2023, EEA, Pompeu Fabra University, Spain.


@Misc{Ilinov_EEA2023,
  author = {Pavel Ilinov and Andrei Matveenko and Salil Sharma},
  title = {Is it better to be first? Search with endogenous information},
  howpublished = {\url{https://transp-or.epfl.ch/documents/talks/Ilinov_EEA2023.pdf}},
  year = {2023},
  note = {Presented at EEA-ESEM 2023, EEA, Pompeu Fabra University, Spain}}}

Ilinov P. Is it better to be first? June 16, 2023. 13th Conference on Economic Design, Society for Economic Design, University of Girona, Spain.


@Misc{Ilinov_ED2023,
  author = {Pavel Ilinov},
  title = {Is it better to be first? },
  howpublished = {\url{https://transp-or.epfl.ch/documents/talks/Ilinov_ED2023.pdf}},
  year = {2023},
  note = {Presented at 13th Conference on Economic Design, Society for Economic Design, University of Girona, Spain}}}

Ilinov P. Is it better to be first? June 09, 2023. Oligo Workshop 2023, Oligo Society, University of Padova, Italy.


@Misc{Ilinov_OW2023,
  author = {Pavel Ilinov},
  title = {Is it better to be first? },
  howpublished = {\url{https://transp-or.epfl.ch/documents/talks/Ilinov_OW2023.pdf}},
  year = {2023},
  note = {Presented at Oligo Workshop 2023, Oligo Society, University of Padova, Italy}}}

Ilinov P. Is it better to be first? June 02, 2023. 15th workshop on Discrete Choice Models, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland.


@Misc{Ilinov_DC2023,
  author = {Pavel Ilinov},
  title = {Is it better to be first? },
  howpublished = {\url{https://transp-or.epfl.ch/documents/talks/Ilinov_DC2023.pdf}},
  year = {2023},
  note = {Presented at 15th workshop on Discrete Choice Models, Ecole Polytechnique F\éd\érale de Lausanne, Lausanne, Switzerland}}}

Reviewing [1]

Journal of Choice Modelling (1)

EPFL courses [3]

Spring 2025

Optimization and Simulation.
- Level: PhD
- Section: Doctoral program in Civil and Environmental Engineering
- Lecturers: Michel Bierlaire, Cloe Cortes Balcells, Lea Ricard, Pavel Ilinov
- Webpage: http://transp-or.epfl.ch/courses/OptSim2025

Fall 2024

Mathematical modeling of behavior.
- Level: Master
- Sections: Mathematics, Master in Financial Engineering
- Lecturer: Michel Bierlaire
- Teaching assistants: Evangelos Paschalidis, Negar Rezvany, Cloe Cortes Balcells, Pavel Ilinov
- Webpage: https://moodle.epfl.ch/course/view.php?id=1001

Spring 2024

Optimization and Simulation.
- Level: PhD
- Section: Doctoral program in Civil and Environmental Engineering
- Lecturers: Michel Bierlaire, Cloe Cortes Balcells, Lea Ricard, Pavel Ilinov
- Webpage: http://transp-or.epfl.ch/courses/OptSim2024

Lectures [3]

2025

Hybrid Choice Models.
- Discrete Choice Analysis: Predicting Individual Behavior and Market Demand, EPFL, Lausanne, Switzerland
- Jan 29, 2025
- Lecturer: Cloe Cortes Balcells
- Assistants: Negar Rezvany, Pavel Ilinov

2024

Bayesian Estimation; Individual Prediction.
- Discrete Choice Analysis: Bayesian estimation and Individual prediction, EPFL, Lausanne, Switzerland
- Feb 15, 2024
- Lecturer: Pavel Ilinov
- Assistants: Marija Kukic, Evangelos Paschalidis
Nested logit; Aggregate forecasting.
- Discrete Choice Analysis: Predicting Individual Behavior and Market Demand, EPFL, Lausanne, Switzerland
- Feb 12, 2024
- Lecturer: Cloe Cortes Balcells
- Assistants: Pavel Ilinov, Evangelos Paschalidis

MSc theses [2]

2025

Adapting simulation algorithms for generating synthetic panel data at the household level, Arturo Cerasi (Computer science)
- Supervision: Marija Kukic, Pavel Ilinov, Michel Bierlaire
- External expert:

2024

Application of the information-theoretic clustering to the discrete choice, Alexandre Monti (Mathematics)
- Supervision: Pavel Ilinov, Evangelos Paschalidis, Michel Bierlaire
- External expert:

Semester projects [1]

2025

Gibbs Sampler for Generating Synthetic Longitudinal Data: Tracking Synthetic Individuals Over Time, Victoria Catherine Arduini (Computer science)

Supervision: Marija Kukic, Pavel Ilinov, Michel Bierlaire
The aim of this project is to extend the existing framework to account for a broader range of attributes, such as work location and income. The student will first define and model a universal set of attributes of individuals and establish corresponding event-duration pairs. Next, they will review the literature to identify appropriate prior distributions for generating each attribute. However, in the event of a disaster, standard priors may fail to accurately model the distributions. To address this, we aim to explore how priors can be adjusted in such cases (e.g., during COVID-19) by incorporating real-world data from the Swiss Mobility and Transport Microcensus (MTMC) \citep{mtmcjedan}. The student will then compare synthetic population generation using only theoretical priors versus updated posteriors. Finally, the student will validate the approach, demonstrating that the impact of a disaster is properly reflected in the generated data.

Michel Bierlaire, EPFL. Sun Mar 30 18:50:32 2025