Seventh international online seminar "Artificial Societies and Information Technologies"

Data-Driven Simulation Calibration with Machine Learning
Tuesday, December 22, 2020
Il-Chul Moon

On December 22, 2020, a meeting of the international online seminar "Artificial Societies and Information Technologies" was held, dedicated to the report Il-Chul Moon (PhD in Computation) "Data-Driven Simulation Calibration with Machine Learning".

Brief annotation.

Agent based models (ABM) has widely utilized because of intuitive analogy to the real world and its flexible expressionism. However, this intuition and the flexibility is gained with the cost of many parameters and difficulties in calibrating them. This talk presents a recent approach of calibrating and validating an ABM model with machine learning approaches. We may consider an ABM as a generator to produce a sample trajectory of the simulated system. Then, our question becomes calibrating the generator from the likelihood-free aspect. We show that our case study of calibration in the housing market and the city-scale transportation.

The meeting was attended by representatives of Russia (CEMI RAS, Moscow State University, Ufa Scientific Center RAS, Ufa State Aviation Technical University, Institute of Economics KarRC RAS, Voronezh State University, Orel State Technical University), China (Shanghai Tianjin Company), India (National Institute of Technology Durgapur), Kazakhstan (L.N. Gumilyov Eurasian National University, Nur-Sultan).

Video of this seminar on our YouTube channel