Wheat stripe rust,caused by Puccinia striiformis f.sp.tritici(Pst),is one of the most destructive fungal diseases of wheat,and seriously threatens safe production of the crop worldwide.In China,new races historically ...Wheat stripe rust,caused by Puccinia striiformis f.sp.tritici(Pst),is one of the most destructive fungal diseases of wheat,and seriously threatens safe production of the crop worldwide.In China,new races historically appeared and rapidly developed to be predominant races and have resulted in ineffectiveness and replacement of wheat resistance cultivars as well as massive reduction in yield.In the present study,the relative parasitic fitness of the two newlyemerged Yr5-virulent races(TSA-6 and TSA-9)were compared with those of four currently predominant Chinese races(CYR31,CYR32,CYR33,and CYR34)based on evaluation on 10 Chinese wheat cultivars.As a result,there were significant differences in the relative parasitic fitness parameters among overall tested races based on multiple comparison(LSD)analysis(P<0.05).The principal component analysis(PCA)of overall parasitic fitness parameters indicated that the sporulation ability,infection and spore survivability,expansion capacity,and potential pathogenicity were the most important parasitic fitness attributes of the tested races.Based on the establishment of extracted three principal components and a comprehensive factor score mathematical models,evaluations of the parasitic fitness attributes of tested races showed that the level of relative parasitic fitness of the tested six races was:CYR32(1.15)>TSA-9(0.95)>TSA-6(0.92)>CYR34(0.29)>CYR31(–1.54)>CYR33(–1.77).The results indicated that two Yr5-virulent races TSA-9 and TSA-6 possessed relative parasitic fitness higher than races CYR34,CYR31,and CYR33,but lower than race CYR32,and have potential risks in developing to be predominant races.Therefore,continual monitoring of both Yr5-virulent races,and their variants is needed.The use of wheat cultivars(lines)with Yr5 resistance gene singly in wheat breeding is essential for being avoided,and is suggested to combine with other effective stripe rust resistance genes.展开更多
无人驾驶技术的关键是决策层根据感知环节输入信息做出准确指令。强化学习和模仿学习比传统规则更适用于复杂场景。但以行为克隆为代表的模仿学习存在复合误差问题,使用优先经验回放算法对行为克隆进行改进,提升模型对演示数据集的拟合...无人驾驶技术的关键是决策层根据感知环节输入信息做出准确指令。强化学习和模仿学习比传统规则更适用于复杂场景。但以行为克隆为代表的模仿学习存在复合误差问题,使用优先经验回放算法对行为克隆进行改进,提升模型对演示数据集的拟合能力;原DDPG(deep deterministic policy gradient)算法存在探索效率低下问题,使用经验池分离以及随机网络蒸馏技术(random network distillation,RND)对DDPG算法进行改进,提升DDPG算法训练效率。使用改进后的算法进行联合训练,减少DDPG训练前期的无用探索。通过TORCS(the open racing car simulator)仿真平台验证,实验结果表明该方法在相同的训练次数内,能够探索出更稳定的道路保持、速度保持和避障能力。展开更多
基金supported by the National Natural Science Foundation of China(32072358 and 32272507)the National Key R&D Program of China(2021YFD1401000)+1 种基金the earmarked fund for CARS-03,the Natural Science Basic Research Project in Shaanxi Province of China(2020JZ-15)National“111 Project”of China(BP0719026)。
文摘Wheat stripe rust,caused by Puccinia striiformis f.sp.tritici(Pst),is one of the most destructive fungal diseases of wheat,and seriously threatens safe production of the crop worldwide.In China,new races historically appeared and rapidly developed to be predominant races and have resulted in ineffectiveness and replacement of wheat resistance cultivars as well as massive reduction in yield.In the present study,the relative parasitic fitness of the two newlyemerged Yr5-virulent races(TSA-6 and TSA-9)were compared with those of four currently predominant Chinese races(CYR31,CYR32,CYR33,and CYR34)based on evaluation on 10 Chinese wheat cultivars.As a result,there were significant differences in the relative parasitic fitness parameters among overall tested races based on multiple comparison(LSD)analysis(P<0.05).The principal component analysis(PCA)of overall parasitic fitness parameters indicated that the sporulation ability,infection and spore survivability,expansion capacity,and potential pathogenicity were the most important parasitic fitness attributes of the tested races.Based on the establishment of extracted three principal components and a comprehensive factor score mathematical models,evaluations of the parasitic fitness attributes of tested races showed that the level of relative parasitic fitness of the tested six races was:CYR32(1.15)>TSA-9(0.95)>TSA-6(0.92)>CYR34(0.29)>CYR31(–1.54)>CYR33(–1.77).The results indicated that two Yr5-virulent races TSA-9 and TSA-6 possessed relative parasitic fitness higher than races CYR34,CYR31,and CYR33,but lower than race CYR32,and have potential risks in developing to be predominant races.Therefore,continual monitoring of both Yr5-virulent races,and their variants is needed.The use of wheat cultivars(lines)with Yr5 resistance gene singly in wheat breeding is essential for being avoided,and is suggested to combine with other effective stripe rust resistance genes.
文摘无人驾驶技术的关键是决策层根据感知环节输入信息做出准确指令。强化学习和模仿学习比传统规则更适用于复杂场景。但以行为克隆为代表的模仿学习存在复合误差问题,使用优先经验回放算法对行为克隆进行改进,提升模型对演示数据集的拟合能力;原DDPG(deep deterministic policy gradient)算法存在探索效率低下问题,使用经验池分离以及随机网络蒸馏技术(random network distillation,RND)对DDPG算法进行改进,提升DDPG算法训练效率。使用改进后的算法进行联合训练,减少DDPG训练前期的无用探索。通过TORCS(the open racing car simulator)仿真平台验证,实验结果表明该方法在相同的训练次数内,能够探索出更稳定的道路保持、速度保持和避障能力。
