正則化多任務(wù)學習的快速算法
發(fā)布時間:2018-09-13 15:34
【摘要】:正則化多任務(wù)學習(regularized multi-task learning,r MTL)方法及其擴展方法在理論研究及實際應(yīng)用方面已經(jīng)取得了較好的成果。然而以往方法僅關(guān)注于多個任務(wù)之間的關(guān)聯(lián),而未充分考慮算法的復雜度,較高的計算代價限制了其在大數(shù)據(jù)集上的實用性。針對此不足,結(jié)合核心向量機(core vector machine,CVM)理論,提出了適用于多任務(wù)大數(shù)據(jù)集的快速正則化多任務(wù)學習(fast regularized multi-task learning,Fr MTL)方法。Fr MTL方法有著與r MTL方法相當?shù)姆诸愋阅?而基于CVM理論的Fr MTL-CVM算法的漸近線性時間復雜度又能使其在面對大數(shù)據(jù)集時仍然能夠獲得較快的決策速度。該方法的有效性在實驗中得到了驗證。
[Abstract]:Regularized multitasking learning (regularized multi-task learning,r MTL) method and its extension method have achieved good results in theoretical research and practical application. However, the previous methods only focus on the correlation between multiple tasks without fully considering the complexity of the algorithm. The high computational cost limits the practicability of the algorithm on big data set. Based on the kernel vector machine (core vector machine,CVM) theory, a fast regularization multitasking learning (fast regularized multi-task learning,Fr MTL (fast regularized multi-task learning,Fr MTL) method for multitasking big data set is proposed. The Fr MTL method has the same classification performance as the r MTL method. However, the asymptote linear time complexity of Fr MTL-CVM algorithm based on CVM theory can make it obtain faster decision speed when facing big data set. The effectiveness of the method is verified by experiments.
【作者單位】: 江南大學數(shù)字媒體學院;無錫職業(yè)技術(shù)學院物聯(lián)網(wǎng)學院;
【基金】:國家自然科學基金面上項目No.61170122 教育部新世紀優(yōu)秀人才支持計劃No.NCET-120882 江蘇省高校品牌專業(yè)建設(shè)工程資助項目No.PPZY2015C240~~
【分類號】:TP18
本文編號:2241588
[Abstract]:Regularized multitasking learning (regularized multi-task learning,r MTL) method and its extension method have achieved good results in theoretical research and practical application. However, the previous methods only focus on the correlation between multiple tasks without fully considering the complexity of the algorithm. The high computational cost limits the practicability of the algorithm on big data set. Based on the kernel vector machine (core vector machine,CVM) theory, a fast regularization multitasking learning (fast regularized multi-task learning,Fr MTL (fast regularized multi-task learning,Fr MTL) method for multitasking big data set is proposed. The Fr MTL method has the same classification performance as the r MTL method. However, the asymptote linear time complexity of Fr MTL-CVM algorithm based on CVM theory can make it obtain faster decision speed when facing big data set. The effectiveness of the method is verified by experiments.
【作者單位】: 江南大學數(shù)字媒體學院;無錫職業(yè)技術(shù)學院物聯(lián)網(wǎng)學院;
【基金】:國家自然科學基金面上項目No.61170122 教育部新世紀優(yōu)秀人才支持計劃No.NCET-120882 江蘇省高校品牌專業(yè)建設(shè)工程資助項目No.PPZY2015C240~~
【分類號】:TP18
【相似文獻】
相關(guān)期刊論文 前1條
1 毛文濤;王海成;劉尚旺;;可自適應(yīng)分組的超球多任務(wù)學習算法[J];計算機應(yīng)用;2014年07期
相關(guān)博士學位論文 前1條
1 浦劍;多任務(wù)學習算法研究[D];復旦大學;2013年
相關(guān)碩士學位論文 前2條
1 黃艾青;基于疊加模型的多任務(wù)學習及應(yīng)用[D];中國科學技術(shù)大學;2015年
2 金鳳;多視角的構(gòu)建及其在單任務(wù)學習和多任務(wù)學習中的應(yīng)用[D];華東師范大學;2010年
,本文編號:2241588
本文鏈接:http://sikaile.net/kejilunwen/zidonghuakongzhilunwen/2241588.html
最近更新
教材專著