報(bào) 告 人：王光臣
報(bào)告題目：A kind of LQ differential game driven by stochastic large-population system
報(bào)告時(shí)間：2025年11月28日（星期五）上午10：40-12：00
報(bào)告人簡(jiǎn)介：王光臣，山東大學(xué)控制科學(xué)與工程學(xué)院常務(wù)副院長(zhǎng)、教授、博導(dǎo)，國(guó)家杰出青年基金獲得者（2019）。一直從事基于不完備信息的隨機(jī)系統(tǒng)優(yōu)化控制理論研究，取得了以倒向分離原理為特色的創(chuàng)新成果，在Springer出版社出版英文專著1部，在控制理論國(guó)際期刊IEEE TAC、Automatica和SIAM J. Control Optim.發(fā)表學(xué)術(shù)論文多篇，曾獲得教育部高等學(xué)校自然科學(xué)獎(jiǎng)二等獎(jiǎng)(第1位)和山東省自然科學(xué)獎(jiǎng)一等獎(jiǎng)（第2位）。 目前主持國(guó)家重點(diǎn)研發(fā)計(jì)劃重點(diǎn)專項(xiàng)。
報(bào)告摘要：This talk is concerned with an LQ differential game driven by partial information stochastic large-population system, in which the large-population system satisfies a backward rather than forward SDE, and both coupling structure and solution of the backward SDE enter state equation and cost functional. Combining maximum principle with optimal filtering, we derive an optimal control of a limiting control problem first, and then, we verify that a decentralized control strategy is an epsilon-Nash equilibrium point of the game. We also work out a numerical example to illustrate the results obtained above.