一、項(xiàng)目詳情
  本項(xiàng)目為學(xué)生提供在監(jiān)督學(xué)習(xí)環(huán)境中所需的現(xiàn)代數(shù)學(xué)工具。我們將重點(diǎn)分析和設(shè)計(jì)正則化方法以及一階優(yōu)化方法，包括隨機(jī)梯度下降法，并介紹統(tǒng)計(jì)學(xué)理論中的主要思想。學(xué)生將在項(xiàng)目結(jié)束時(shí)，提交項(xiàng)目研究報(bào)告，進(jìn)行成果展示。
  機(jī)器學(xué)習(xí)算法優(yōu)化背后的計(jì)算數(shù)學(xué)理論與應(yīng)用研究培訓(xùn)
  二、適合人群
  大學(xué)生
  希望申請(qǐng)人工智能，數(shù)據(jù)科學(xué)，統(tǒng)計(jì)學(xué)等相關(guān)領(lǐng)域博士的學(xué)生，學(xué)生需要具備微積分、線性代數(shù)、概率論基礎(chǔ)，至少會(huì)使用一門(mén)編程語(yǔ)言實(shí)現(xiàn)如KNN等經(jīng)典機(jī)器學(xué)習(xí)算法
  三、項(xiàng)目大綱
  統(tǒng)計(jì)學(xué)習(xí)：監(jiān)督學(xué)習(xí)和無(wú)監(jiān)督學(xué)習(xí)、建模數(shù)據(jù)、多元正態(tài)模型、正態(tài)線性模型、貝葉斯學(xué)習(xí)Statistical Learning.Topics in this week cover supervised and unsupervised learning,modeling data,multivariate normal models,normal linear models,and Bayesian learning.
  回歸理論：線性回歸、線性模型分析、模型選擇與預(yù)測(cè)、非線性回歸、線性模型Regression.Topics in this week are about linear regression,analysis of linear models,model selection and prediction,nonlinear regression,and linear models in Python.
  核與正則化方法:正則化、表示定理、高斯回歸、核主成分分析Regularization and Kernel methods.Topics in this week cover regularization,re-presenter theorem,Gaussian process regression,and Kernel principal component analysis.
  機(jī)器學(xué)習(xí)優(yōu)化方法Optimization for machine learning
  分類(lèi)、決策樹(shù)和集成方法:支持向量機(jī)、決策樹(shù)、隨機(jī)森林Classification,decision trees and ensemble methods.Topics in this week include support vector machine,decision trees,and random forests.
  項(xiàng)目回顧與成果展示Program review and presentation
  論文輔導(dǎo)和發(fā)表Project deliverables tutoring
  模擬博士申請(qǐng)面試Mock interview for PhD application
  四、時(shí)間安排與收獲
  7周在線小組科研學(xué)習(xí)+5周論文輔導(dǎo)學(xué)習(xí)共125課時(shí)
  學(xué)術(shù)報(bào)告
  *學(xué)員獲主導(dǎo)師Reference Letter
  EI/CPCI/Scopus/ProQuest/Crossref/EBSCO或同等級(jí)別索引國(guó)際會(huì)議全文投遞與發(fā)表（可用于申請(qǐng)）
  結(jié)業(yè)證書(shū)
  成績(jī)單