检索结果-内蒙古大学图书馆

IEEE Conference on Games (CoG)

作者： Huang, Shiyu Yu, Chao Wang, Bin Li, Dong Wang, Yu Chen, Ting Zhu, Jun Tsinghua Univ Beijing Peoples R China Huawei Noahs Ark Lab Beijing Peoples R China

ISBN: (数字)9781665459891

ISBN: (纸本)9781665459891

Recent algorithms designed for multi-agent tasks focus on finding a single optimal solution for all the agents. However, in many tasks (e.g., matrix games and transportation dispatching), there may exist more than one optimal solution, while previous algorithms can only converge to one of them. In many practical applications, it is important to develop reasonable agents with diverse behaviors. In this paper, we propose "variational multi-agent policy diversification" (VMAPD), an on-policy framework for discovering diverse policies for coordination patterns of multiple agents. By taking advantage of latent variables and exploiting the connection between variational inference and multi-agent reinforcement learning, we derive a tractable evidence lower bound (ELBO) on the trajectories of all agents. Our algorithm uses policy iteration to maximize the derived lower bound and can be simply implemented by adding a pseudo reward during centralized learning. And the trained agents do not need to access the pseudo reward during decentralized execution. We demonstrate the effectiveness of our algorithm on several popular multi-agent testbeds. Experimental results show that VMAPD finds more solutions with similar sample complexity compared with other baselines.

关键词： deep reinforcement learning multi-agent reinforcement learning diversity probabilistic graphical models

来源：评论

学校读者我要写书评

暂无评论

A feature-based approach to modeling protein-DNA interactions

A feature-based approach to modeling protein-DNA interaction...

引用

11th Annual International Conference on Research in Computational Molecular Biology

作者： Sharon, Eilon Segal, Eran Weizmann Inst Sci Dept Comp Sci IL-76100 Rehovot Israel

ISBN: (纸本)9783540716808

Transcription factor (TF) binding to its DNA target site is a fundamental regulatory interaction. The most common model used to represent TF binding specificities is a position specific scoring matriX (PSSM), which assumes independence between binding positions. In many cases this simplifying assumption does not hold. Here, we present feature motif models (FMMs), a novel probabilistic method for modeling TF-DNA interactions, based on Markov networks. Our approach uses sequence features to represent TF binding specificities, where each feature may span multiple positions. We develop the mathematical formulation of our models, and devise an algorithm for learning their structural features from binding site data. We evaluate our approach on synthetic data, and then apply it to binding site and ChIP-chip data from yeast. We reveal sequence features that are present in the binding specificities of yeast TFs, and show that FMMs explain the binding data significantly better than PSSMs.

关键词： transcription factor binding sites DNA sequence motifs probabilistic graphical models Markov networks motif finder

来源：评论

学校读者我要写书评

暂无评论

Microblog Sentiment Topic Model 16

Microblog Sentiment Topic Model

引用

16th IEEE International Conference on Data Mining (ICDM)

作者： Ahuja, Aman Wei, Wei Carley, Kathleen M. BITS Pilani KK Birla Goa Campus Zuarinagar Goa India Carnegie Mellon Univ Pittsburgh PA 15213 USA

ISBN: (纸本)9781509059102

With the prevalence of social media, such as Twitter, short-length text like microblogs have become an important mode of text on the Internet. In contrast to other forms of media, such as newspaper, the text in these social media posts usually contains fewer words, and is concentrated on a much narrower selection of topics. For these reasons, traditional LDA-based sentiment and topic modeling techniques generally do not work well in case of social media data. Another characteristic feature of this data is the use of special meta tokens, such as hashtags, which contain unique semantic meanings that are not captured by other ordinary words. In the recent years, many topic modeling techniques have been proposed for social media data, but the majority of this work does not take into account the specialty of tokens, such as hashtags, and treats them as ordinary words. In this paper, we propose probabilistic graphical models to address the problem of discovering latent topics and their sentiment from social media data, mainly microblogs like Twitter. We first propose MTM (Microblog Topic Model), a generative model that assumes each social media post generates from a single topic, and models both words and hashtags separately. We then propose MSTM (Microblog Sentiment Topic Model), an extension of MTM, which also embodies the sentiment associated with the topics. We evaluated our models using Twitter dataset, and experimental

关键词： probabilistic graphical models Sentiment Analysis Topic Modeling

来源：评论

学校读者我要写书评

暂无评论

Monocular Depth Estimation Based on a Single Image - A Literature Review 12

Monocular Depth Estimation Based on a Single Image - A Liter...

引用

12th International Conference on Graphics and Image Processing (ICGIP)

作者： Tian Yuan Hu Xiaodong Chinese Acad Sci Xian Inst Opt & Precis Mech Beijing Peoples R China Univ Chinese Acad Sci Beijing Peoples R China

ISBN: (纸本)9781510642782;9781510642775

Monocular depth estimation is a very valuable but also very challenging problem. In order to solve this ill-posed problem, traditional approaches apply depth cues such as defocusing, atmospheric scattering and shading to estimate depth information and approaches based on machine learning apply frameworks such as MRF and data-driven learning. With the development of deep learning, the monocular depth estimation approaches based on CNN and other networks have achieved good results and gradually become the mainstream. In this paper, we summarize some typical and representative literature on monocular depth estimation based on a single image in the past two decades and depict our analysis involved in these approaches. In addition, this paper also analyzes and compares the results obtained by some typical approaches, which may provide some guidance for those who are interested in this field.

关键词： Depth estimation depth cues probabilistic graphical models data-driven CNN

来源：评论

学校读者我要写书评

暂无评论

Behaviour Recognition on Noisy Data-streams Constrained by Complex Prior Knowledge

Behaviour Recognition on Noisy Data-streams Constrained by C...

引用

作者： Romain Rince UNIVERSITE DE NANTES

学位级别：博士

Complex Event Processing (CEP) The first one is based on Markov logical consists of the analysis of data-streams in order networks to represent the structure of the to extract particular patterns and behaviours chronicles under a set of logical formulas of a described, in general, in a logical formalism. In confidence value. We show that this model, the classical approach, data of a stream -- or although widely applied in the literature, is events -- are supposed to be the complete and inapplicable for a realistic application with perfect observation of the system producing regard to the dimensions of such a problem. these events. However, in many cases, the The second approach is based on techniques means for collecting such data, such as from the SAT community to enumerate all sensors, are not infallible and may miss the possible solutions ofa given problem and thus detection of a particular event or on the contrary to produce a confidence value for the produce. In this thesis, we have studied the recognition of a chronicle expressed, again, possible models of representation of uncertainty under a logical structure. Finally, we propose a and, thus, to offer the CEP a robustness to this last approach based on the Markov chains to uncertainty as well as the necessary tools to produce a set of samples explaining the allow the recognition of complex behaviours evolution of the model in agreement with the based on the chronicle formalism. In this observed data. These samples are then perspective, three approaches have been analysed by a recognition system to count the considered. occurrences of a particular chronicle.

关键词： Complex event processing incertitude representation Markov logic networks Non-homogenous Markov models probabilistic graphical models

来源：评论

学校读者我要写书评

暂无评论

Stokes, Gauss, and Bayes Walk into a Bar ...

Stokes, Gauss, and Bayes Walk into a Bar ...

引用

作者： Kightley, E.P. University of Colorado at Boulder

学位级别：Ph.D.

This thesis consists of three distinct projects. The first is a study of microbial aggregate fragmentation, in which we develop a dynamical model of aggregate deformation and breakage and use it to obtain a post-fragmentation density function. The second and third projects deal with dimensionality reduction in machine learning problems. In the second project, we derive a one-pass sparsified Gaussian mixture model to perform clustering analysis on high-dimensional streaming data. The model estimates parameters in dense space while storing and performing computations in a compressed space. In the final project, we build an expert system classifier with a Bayesian network for use on high-volume streaming data. Our approach is specialized to reduce the number of observations while obtaining sufficient labeled training data in a regime of extreme class-imbalance and expensive oracle queries.

关键词： microbial flocculation dimensionality reduction random projections Gaussian mixture models clustering Bayesian networks probabilistic graphical models anomaly detection classification

来源：评论

学校读者我要写书评

暂无评论

Computing Marginal and Conditional Divergences between Decomposable models with Applications 23

Computing Marginal and Conditional Divergences between Decom...

引用

23rd IEEE International Conference on Data Mining (IEEE ICDM)

作者： Lee, Loong Kuan Webb, Geoffrey I. Schmidt, Daniel F. Piatkowski, Nico Fraunhofer IAIS Schloss Birlinghoven D-53757 St Augustin Germany Monash Univ Dept Data Sci & AI Melbourne Vic Australia

ISBN: (纸本)9798350307887

The ability to compute the exact divergence between two high-dimensional distributions is useful in many applications but doing so naively is intractable. Computing the alpha-beta divergence a family of divergences that includes the Kullback-Leibler divergence and Hellinger distance between the joint. distribution of two decomposable models, i.e chordal Markov networks, can be done in time exponential in the treewidth of these models. However, reducing the dissimilarity between two high-dimensional objects to a single scalar value can be uninformative. Furthermore, in applications such as supervised learning, the divergence over a conditional distribution might be of more interest. Therefore, we propose an approach to compute the exact alpha-beta divergence between any marginal or conditional distribution of two decomposable models. Doing so tractably is non -trivial as we need to decompose the divergence between these distributions and therefore, require a decomposition over the marginal and conditional distributions of these models. Consequently, we provide such a decomposition and also extend existing work to compute the marginal and conditional alpha beta divergence between these decompositions. We then show how our method can be used to analyze distributional changes by first applying it to a benchmark image dataset. Finally, based on our framework, we propose a novel way to quantify the error in contemporary superconducting quantum computers.

关键词： alpha-beta divergence Bayesian network decomposable models divergences Hellinger distance Kullback-Leibler divergence Markov network probabilistic graphical models

来源：评论

学校读者我要写书评

暂无评论

A Generic Framework for Behavior Recognition of Complex Activities in Robotics

A Generic Framework for Behavior Recognition of Complex Acti...

引用

25th IEEE International Conference on Tools with Artificial Intelligence (ICTAI)

作者： Haeussermann, Kai Zweigle, Oliver Levi, Paul Univ Stuttgart Dept Image Understanding D-70469 Stuttgart Germany

ISBN: (纸本)9781479929719

Considering the current state in service-robotics, an expert is still necessary to add new tasks and execution behaviors by textual and error-prone programming. Under the consideration that humans typically execute same activities almost identical (or at least similar) and further combine simple behaviors to more complex activities, we follow the constitutive assumption that all complex behaviors are composed of a limited set of atomic behaviors. This work introduces a generic framework for spatial-temporal analysis and classification of arbitrary atomic behaviors. Therefore, we propose the combination of Self-Organizing Maps (SOM) and probabilistic graphical models (PGM) in order to exploit the advantages of both concepts. In this work, we describe the essential methods of the framework briefly, whereas the data-driven training of the spatial-temporal model and the reasoning process are described in detail. In order to demonstrate the potential and to emphasize the high level of generalization and flexibility in real-world environments, the framework is evaluated in an exemplary scenario.

关键词： Self-Organizing Maps probabilistic graphical models Behavior Recognition

来源：评论

学校读者我要写书评

暂无评论

A GENERAL BAYESIAN ALGORITHM FOR VISUAL OBJECT TRACKING BASED ON SPARSE FEATURES

A GENERAL BAYESIAN ALGORITHM FOR VISUAL OBJECT TRACKING BASE...

引用

IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Soto, Mauricio Regazzoni, Carlo S. Univ Genoa Dept Biophys & Elect Engn Genoa Italy

ISBN: (纸本)9781457705397

This paper describes a Bayesian algorithm for rigid/non-rigid 2D visual object tracking based on sparse image features. The algorithm is inspired by the way human visual cortex segments and tracks different moving objects within its FOV by constructing dynamical nonretinotopic layers. The method is explained as a recursive algorithm between time slices (intra-slice) and as a forward-backward message passing within every time slice (inter-slice) under the probabilistic graphical Model (PGM) framework. Finally, an observation model function that resembles the Generalized Hough Transform and allows exploiting internal structure of the problem is employed in order to increase the robustness and accuracy of the algorithm against clutter and missed detections.

关键词： Visual Object Tracking probabilistic graphical models Nonretinotopic Representation Generalized Hough Transform

来源：评论

学校读者我要写书评

暂无评论

Tutorial of the probabilistic methods Bayesian networks and influence diagrams applied to medicine

引用

Journal of Evidence-Based Medicine 2018年第2期11卷 112-124页

作者： Nistal-Nuño, Beatriz Department of Anesthesiology Yale University School of Medicine New Haven CT United States

Unpredictability arises in numerous real-world implementations especially in medical fields, being one of the principal subjects that expert systems have to deal with. Bayesian networks (BNs) allow constructing expert systems by utilizing likelihood as a measure of uncertainty. A BN is a statistical model that yields a graphical description of the probabilistic associations between a group of pertinent variables. probabilistic reasoning involves computing the posterior probability of the unobserved variables given the accessible evidence, based on the application of the Bayes theorem. The principal benefit from BNs is that intricate relationships are decreased to simpler local dependencies. graphical models utilization in health economics and medical decision-making has been supported lately. The non-existence of objective data frequently pressures the knowledge engineer to acquire the likelihoods from human experts’ estimations, which turns this task laborious. For this logic, a software explanation facility can help the expert and the knowledge engineer through this procedure. In order to depict graphically and approximate a decision problem, a BN can be extended with decision and utility nodes associated with the realization of the random variables. Such model is named influence diagram (ID). The main aim of this article is to present a tutorial of BNs and the decision-theoretic model's IDs applied to medicine, including an explanation of these methods through several theoretical examples and illustrations of studies in the medical literature. © 2018 Chinese Cochrane Center, West China Hospital of Sichuan University and John Wiley & Sons Australia, Ltd

关键词： Bayesian networks electronic health records influence diagrams probabilistic graphical models

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：