時(shí)間：12月6日（星期一）下午15:00-16:30
騰訊會(huì)議號(hào)：117 564 951
報(bào)告人：西安交通大學(xué)劉佳鵬 副教授
主講人簡(jiǎn)介：
劉佳鵬博士，西安交通大學(xué)偉德國(guó)際1946bv官網(wǎng)智能決策與機(jī)器學(xué)習(xí)研究中心副教授、博士生導(dǎo)師。目前的研究方向包括：決策分析、機(jī)器學(xué)習(xí)、貝葉斯方法、大數(shù)據(jù)模型。主持過國(guó)家自然科學(xué)基金青年項(xiàng)目及面上項(xiàng)目、國(guó)家重點(diǎn)研發(fā)計(jì)劃項(xiàng)目子課題以及博士后科學(xué)基金項(xiàng)目。研究成果發(fā)表在INFORMS Journal on Computing、European Journal of Operational Research、Omega、Knowledge-based Systems、系統(tǒng)工程理論與實(shí)踐、系統(tǒng)工程學(xué)報(bào)等國(guó)內(nèi)外重要學(xué)術(shù)期刊?，F(xiàn)擔(dān)任中國(guó)優(yōu)選法統(tǒng)籌法與經(jīng)濟(jì)數(shù)學(xué)研究會(huì)智能決策與博弈分會(huì)理事、中國(guó)系統(tǒng)工程學(xué)會(huì)數(shù)據(jù)科學(xué)與知識(shí)系統(tǒng)工程專委會(huì)委員。
報(bào)告內(nèi)容簡(jiǎn)介：
We propose a preference learning algorithm for uncovering Decision Makers’(DMs’) contingent evaluation strategies in the context of multiple criteria sorting. We assume the preference information in the form of holistic assignment examples derived from the analysis of alternatives’ performance vectors and textual descriptions. We characterize the decision policies using a mixture of threshold-based value-driven preference models and associated latent topics. The latter serve as the stimuli underlying the contingency in decision behavior, providing a transparent and interpretable way to explore and understand DMs’ contingent preferences. Such a probabilistic model is constructed using a flexible and nonparametric Bayesian framework. The proposed method adopts a hierarchical Dirichlet process so that a group of DMs can share a countably infinite number of contingent models and topics. For all DMs, it automatically identifies the components representing their evaluation strategies adequately. The posterior is summarized using the Hamiltonian Monte Carlo sampling method. We demonstrate the method’s practical usefulness on a real-world recruitment problem considered by a Chinese IT company. We discuss the contingent models and topics and illustrate their employment for classifying the job applicants. We also compare the approach with counterparts that use just a single preference model, implement the parametric framework, or consider each DM’s preferences individually.
（承辦：管理工程系、科研與學(xué)術(shù)交流中心）
