知识推理是补全知识图谱的重要方法,旨在根据图谱中已有的知识,推断出未知的事实或关系.针对多数推理方法仍存在没有充分考虑实体对之间的路径信息,且推理效率偏低、可解释性差的问题,提出了将TuckER嵌入和强化学习相结合的知识推理方法...知识推理是补全知识图谱的重要方法,旨在根据图谱中已有的知识,推断出未知的事实或关系.针对多数推理方法仍存在没有充分考虑实体对之间的路径信息,且推理效率偏低、可解释性差的问题,提出了将TuckER嵌入和强化学习相结合的知识推理方法TuckRL(TuckER embedding with reinforcement learning).首先,通过TuckER嵌入将实体和关系映射到低维向量空间,在知识图谱环境中采用策略引导的强化学习算法对路径推理过程进行建模,然后在路径游走进行动作选择时引入动作修剪机制减少无效动作的干扰,并将LSTM作为记忆组件保存智能体历史动作轨迹,促使智能体更准确地选择有效动作,通过与知识图谱的交互完成知识推理.在3个主流大规模数据集上进行了实验,结果表明TuckRL优于现有的大多数推理方法,说明将嵌入和强化学习相结合的方法用于知识推理的有效性.展开更多
This paper concerns the observer-based adaptive control problem of uncertain time-delay switched systems with stuck actuator faults. Under the case where the original controller cannot stabilize the faulty system, mul...This paper concerns the observer-based adaptive control problem of uncertain time-delay switched systems with stuck actuator faults. Under the case where the original controller cannot stabilize the faulty system, multiple adaptive controllers are designed and a suitable switching logic is incorporated to ensure the closed-loop system stability and state tracking. New delay-independent sufficient conditions for asymptotic stability are obtained in terms of linear matrix inequalities based on piecewise Lyapunov stability theory. On the other hand, adaptive laws for on-line updating of some of the controller parameters are also designed to compensate the effect of stuck failures. Finally, simulation results for reference [1] model show that the design is feasible and efficient.展开更多
Purpose:.To evaluate the surgical effect of levator muscle shortening and levator aponeurosis tucking in treating minimal and moderate congenital blepharoptosis.Methods:.Clinical data of 28 patients(40 eyes) diagnosed...Purpose:.To evaluate the surgical effect of levator muscle shortening and levator aponeurosis tucking in treating minimal and moderate congenital blepharoptosis.Methods:.Clinical data of 28 patients(40 eyes) diagnosed with mide and moderate congenital blepharoptosis at our institution were retrospectively analyzed. Postoperative efficacy was evaluated and statistically compared between these two techniques.Results:.During 14 months follow-up,.16 eyes with ptosis undergoing levator muscle shortening were treated,.3 with undercorrection of ptosis and 1 with overcorrection of ptosis.In patients receiving levator aponeurosis tucking,.16 eyes were cured and 4 with undercorrection of ptosis.Conclusion:.Both levator muscle shortening and levator aponeurosis tucking are safe and efficacious for correcting minimal and moderate congenital blepharoptosis.展开更多
低秩稀疏分解方法因其好的检测性能在红外小目标检测领域受到广泛关注。然而,现有低秩稀疏分解方法在复杂场景中仍然面临检测性能不高、检测速度较慢等问题。虽然现有的低秩塔克分解方法在复杂场景下取得了令人满意的检测性能,但其需依...低秩稀疏分解方法因其好的检测性能在红外小目标检测领域受到广泛关注。然而,现有低秩稀疏分解方法在复杂场景中仍然面临检测性能不高、检测速度较慢等问题。虽然现有的低秩塔克分解方法在复杂场景下取得了令人满意的检测性能,但其需依赖经验预先定义秩:若秩估计过大或过小,会导致漏检或虚警。而且,不同场景中秩的大小不一样,限制了实际应用。为了解决这一问题,本文采用非凸秩接近范数约束低秩塔克分解的潜在因子,无需手动设置秩,从而显著提升了算法在不同场景中的鲁棒性。进一步地,设计了基于对称高斯-赛德尔的交替方向乘子法(symmetric GaussSeidel based alternating direction method of multipliers algorithm,sGSADMM)来求解所提模型。与现有基于交替方向乘子法相比,sGSADMM算法通过利用更多结构信息,实现了更高的求解精度。大量实验表明,所提方法在检测性能和背景抑制等方面均优于现有的先进算法。展开更多
文摘知识推理是补全知识图谱的重要方法,旨在根据图谱中已有的知识,推断出未知的事实或关系.针对多数推理方法仍存在没有充分考虑实体对之间的路径信息,且推理效率偏低、可解释性差的问题,提出了将TuckER嵌入和强化学习相结合的知识推理方法TuckRL(TuckER embedding with reinforcement learning).首先,通过TuckER嵌入将实体和关系映射到低维向量空间,在知识图谱环境中采用策略引导的强化学习算法对路径推理过程进行建模,然后在路径游走进行动作选择时引入动作修剪机制减少无效动作的干扰,并将LSTM作为记忆组件保存智能体历史动作轨迹,促使智能体更准确地选择有效动作,通过与知识图谱的交互完成知识推理.在3个主流大规模数据集上进行了实验,结果表明TuckRL优于现有的大多数推理方法,说明将嵌入和强化学习相结合的方法用于知识推理的有效性.
基金supported by the National Basic Research Program of China (No.2007CB714006)
文摘This paper concerns the observer-based adaptive control problem of uncertain time-delay switched systems with stuck actuator faults. Under the case where the original controller cannot stabilize the faulty system, multiple adaptive controllers are designed and a suitable switching logic is incorporated to ensure the closed-loop system stability and state tracking. New delay-independent sufficient conditions for asymptotic stability are obtained in terms of linear matrix inequalities based on piecewise Lyapunov stability theory. On the other hand, adaptive laws for on-line updating of some of the controller parameters are also designed to compensate the effect of stuck failures. Finally, simulation results for reference [1] model show that the design is feasible and efficient.
文摘Purpose:.To evaluate the surgical effect of levator muscle shortening and levator aponeurosis tucking in treating minimal and moderate congenital blepharoptosis.Methods:.Clinical data of 28 patients(40 eyes) diagnosed with mide and moderate congenital blepharoptosis at our institution were retrospectively analyzed. Postoperative efficacy was evaluated and statistically compared between these two techniques.Results:.During 14 months follow-up,.16 eyes with ptosis undergoing levator muscle shortening were treated,.3 with undercorrection of ptosis and 1 with overcorrection of ptosis.In patients receiving levator aponeurosis tucking,.16 eyes were cured and 4 with undercorrection of ptosis.Conclusion:.Both levator muscle shortening and levator aponeurosis tucking are safe and efficacious for correcting minimal and moderate congenital blepharoptosis.
文摘低秩稀疏分解方法因其好的检测性能在红外小目标检测领域受到广泛关注。然而,现有低秩稀疏分解方法在复杂场景中仍然面临检测性能不高、检测速度较慢等问题。虽然现有的低秩塔克分解方法在复杂场景下取得了令人满意的检测性能,但其需依赖经验预先定义秩:若秩估计过大或过小,会导致漏检或虚警。而且,不同场景中秩的大小不一样,限制了实际应用。为了解决这一问题,本文采用非凸秩接近范数约束低秩塔克分解的潜在因子,无需手动设置秩,从而显著提升了算法在不同场景中的鲁棒性。进一步地,设计了基于对称高斯-赛德尔的交替方向乘子法(symmetric GaussSeidel based alternating direction method of multipliers algorithm,sGSADMM)来求解所提模型。与现有基于交替方向乘子法相比,sGSADMM算法通过利用更多结构信息,实现了更高的求解精度。大量实验表明,所提方法在检测性能和背景抑制等方面均优于现有的先进算法。
