Structured Learning Team
(https://aip.riken.jp/labs/generic_tech/struct_learn/) at RIKEN AIP
Title: (1) Overview of Structured Learning Team, and (2) Some recent advances in operator-theoretic data analysis for dynamical systems:
My talk consists of the following two parts:
Part 1. When making predictions based on intelligent information processing such as machine learning, we usually have prior information about structures among variables in data. In Structured Learning team, we are have been developing theories and algorithms for learning with such structural information. Also, we have been conducting applied researches by applying developed algorithms to a variety of scientific and engineering data.
Part 2. Data-driven modeling of complex systems has received much attention over the recent years, largely due to the availability of large datasets. Operator-theoretic analysis of nonlinear dynamical systems has been actively discussed in applied mathematics and various scientific fields for this purpose. This is because it can provide physical interpretations of the dynamics based on deep theoretical backgrounds and is endowed with prominent estimation methods such as dynamic mode decomposition (DMD). In this talk, I first overview the recent advances on this research topic, especially focusing on spectral analysis of dynamical systems with Koopman operators and DMD. Then, I describe several recently-proposed related algorithms based on machine learning principles, where I occasionally show some applications of these method to several real-world data.
Title: Data-driven Analysis for Multi-agent Trajectories in Team Sports:
Understanding the principles of real-world biological multi-agent behaviors is a current challenge in various scientific and engineering fields. The rules regarding the real-world biological multi-agent behaviors such as team sports are often largely unknown due to their inherently higher-order interactions, cognition, and body dynamics. Estimation of the rules from multi-agent trajectories, i.e., data-driven approaches using machine learning, provides an effective way for the analysis of such behaviors. In this talk, I mainly introduce two approaches for understanding such multi-agent behaviors: (1) extracting physically-interpretable features of biological network dynamics and (2) generating and controlling behaviors via decentralized policy learning with partial observation and mechanical constraints.
Title: Exploiting variable associations to configure efficient local search algorithms in large-scale binary integer programs:
We present a data mining approach for reducing the search space of local search algorithms in a class of binary integer programs including the set covering and partitioning problems. The quality of locally optimal solutions typically improves if a larger neighborhood is used, while the computation time of searching the neighborhood increases exponentially. To overcome this, we extract variable associations from the instance to be solved in order to identify promising pairs of flipping variables in the neighbor- hood search. Based on this, we develop a 4-flip neighborhood local search algorithm that incorporates an efficient incremental evaluation of solutions and an adaptive control of penalty weights. Computational results show that the proposed method improves the performance of the local search algorithm for large-scale set covering and partitioning problems.
|Date||March 17, 2021 (Wed) 15:00 - 17:00|