• A
  • A
  • A
  • АБB
  • АБB
  • АБB
  • А
  • А
  • А
  • А
  • А
Обычная версия сайта
Важные объявления 1
Уважаемые аспиранты!
Обращаем ваше внимание, что на период сложной эпидемиологической обстановки временно приостановлен очный прием посетителей!
Все возникающие вопросы вы можете адресовать менеджеру по электронной почте.

PhD Research Seminar: Time series and automated machine learning, Truncated Quantile Critics and overestimation bias

Мероприятие завершено
Where: Faculty of Computer Science, Pokrovskii bulvar 11, room R306. 
When: March 2, 19:30–21:00 

First talk: Time Series Forecasting Based on Automated Machine Learning 
Speaker: Konstantin Danilov, second-year PhD student, School of Business Informatics

Nowadays, forecasting time series is important in solving a wide range of problems in various spheres of human activity. Researchers resorted to different approaches to achieve the required accuracy of  forecasting models, including feature engineering. This presentation will be about a feature engineering method for time series data based on Bayesian optimization. The proposed approach has been experimentally shown to be higly efficient. Some difficulties related to this approach will be discussed.

Second talk: Controlling Overestimation Bias with Truncated Mixture of Continuous Distributional Quantile Critics
Speaker: Alexander Grishin, second-year PhD student, Faculty of Computer Science

According to previous studies, one of the major impediments to accurate off-policy learning is the overestimation bias. During the talk, I will explore our novel way to alleviate the overestimation bias in a continuous control setting. Our method, Truncated Quantile Critics (TQC), blends three ideas: distributional representation of a critic, truncation of critics prediction, and ensembling of multiple critics. We show that all components are key for the achieved performance. Distributional representation combined with truncation allows for arbitrary granular overestimation control, and ensembling further improves the results of our method. TQC significantly outperforms the current state of the art on all environments from the continuous control benchmark suite, demonstrating 25% improvement on the most challenging Humanoid environment.