Proximal Policy Optimization (PPO) is a highly popular policy-based deep reinforcement learning (DRL) approach. However, we observe that the homogeneous exploration process in PPO could cause an unexpected stability issue in the training phase. To address this issue, we propose PPO-UE, a PPO variant equipped with self-adaptive uncertainty-aware explorations (UEs) based on a ratio uncertainty level. The proposed PPO-UE is designed to improve convergence speed and performance with an optimized ratio uncertainty level. Through extensive sensitivity analysis by varying the ratio uncertainty level, our proposed PPO-UE considerably outperforms the baseline PPO in Roboschool continuous control tasks.
translated by 谷歌翻译
对不确定性的深入了解是在不确定性下做出有效决策的第一步。深度/机器学习(ML/DL)已被大大利用,以解决处理高维数据所涉及的复杂问题。但是,在ML/DL中,推理和量化不同类型的不确定性的探索少于其他人工智能(AI)领域。特别是,自1960年代以来,在KRR上已经研究了信仰/证据理论,以推理并衡量不确定性以提高决策效率。我们发现,只有少数研究利用了ML/DL中的信念/证据理论中的成熟不确定性研究来解决不同类型的不确定性下的复杂问题。在本调查论文中,我们讨论了一些流行的信念理论及其核心思想,这些理论涉及不确定性原因和类型,并量化它们,并讨论其在ML/DL中的适用性。此外,我们讨论了三种主要方法,这些方法在深度神经网络(DNN)中利用信仰理论,包括证据DNN,模糊DNN和粗糙的DNN,就其不确定性原因,类型和量化方法以及其在多元化问题中的适用性而言。域。根据我们的深入调查,我们讨论了见解,经验教训,对当前最新桥接信念理论和ML/DL的局限性,最后是未来的研究方向。
translated by 谷歌翻译
Recommendation Systems (RSs) are ubiquitous in modern society and are one of the largest points of interaction between humans and AI. Modern RSs are often implemented using deep learning models, which are infamously difficult to interpret. This problem is particularly exasperated in the context of recommendation scenarios, as it erodes the user's trust in the RS. In contrast, the newly introduced Tsetlin Machines (TM) possess some valuable properties due to their inherent interpretability. TMs are still fairly young as a technology. As no RS has been developed for TMs before, it has become necessary to perform some preliminary research regarding the practicality of such a system. In this paper, we develop the first RS based on TMs to evaluate its practicality in this application domain. This paper compares the viability of TMs with other machine learning models prevalent in the field of RS. We train and investigate the performance of the TM compared with a vanilla feed-forward deep learning model. These comparisons are based on model performance, interpretability/explainability, and scalability. Further, we provide some benchmark performance comparisons to similar machine learning solutions relevant to RSs.
translated by 谷歌翻译
我们研究知识图嵌入(KGE)对知识图(KG)完成的有效性,并通过规则挖掘完成。更具体地说,我们在KGE完成之前和之后从KGS中挖掘规则,以比较提取的规则的可能差异。我们将此方法应用于经典的方法,尤其是Transe,Distmult and Complext。我们的实验表明,根据KGE完成的KGE方法,提取的规则之间可能存在巨大差异。特别是,在完成转盘后,提取了几条虚假规则。
translated by 谷歌翻译
时间序列分析中产生的最重要的问题之一是分叉或变化点检测。也就是说,给定时间序列的集合在不同的参数上,何时基础动力系统的结构发生了变化?对于此任务,我们转向拓扑数据分析(TDA)的领域,该领域编码有关数据形状和结构的信息。近年来,利用TDA的工具用于信号处理任务(称为拓扑信号处理(TSP)(TSP))的想法在很大程度上通过标准管道获得了很多关注,该标准管道计算出Takens嵌入产生的点云的持久同源性。但是,此过程受到计算时间的限制,因为在这种情况下生成的简单复合物很大,但也有很多冗余数据。因此,我们求助于编码吸引子结构的最新方法,该方法构建了代表有关何时在状态空间区域之间动态系统传递的信息的序数分区网络(OPN)。结果是一个加权图,其结构编码有关基础吸引子的信息。我们以前的工作开始寻找以TDA适合的方式包装OPN信息的方法。但是,这项工作仅使用网络结构,而没有采取任何行动来编码其他加权信息。在本文中,我们采取下一步:构建管道来分析使用TDA的加权OPN,并表明该框架为系统中的噪声或扰动提供了更大的弹性,并提高了动态状态检测的准确性。
translated by 谷歌翻译