The accumulation of genomic and proteomic data of many organisms presents an opportunity to analyze entire phylogenetic trees in a systematic, quantified manner. The universal tree of life, constructed by genomic data...
详细信息
The least squares approach works efficiently in value function approximation, given appropriate basis functions. Because of its smoothness, the Gaussian kernel is a popular and useful choice as a basis function. Howev...
详细信息
The least squares approach works efficiently in value function approximation, given appropriate basis functions. Because of its smoothness, the Gaussian kernel is a popular and useful choice as a basis function. However, it does not allow for discontinuity which typically arises in real-world reinforcement learning tasks. In this paper, we propose a new basis function based on geodesic Gaussian kernels, which exploits the non-linear manifold structure induced by the Markov decision processes. The usefulness of the proposed method is successfully demonstrated in a simulated robot arm control and Khepera robot navigation.
With myriad information being generated from high-throughput experiments such as microarrays and sequencing technologies, an ever-increasing amount of data is being recorded and analyzed with the help of hierarchical ...
详细信息
ISBN:
(纸本)9780615153148
With myriad information being generated from high-throughput experiments such as microarrays and sequencing technologies, an ever-increasing amount of data is being recorded and analyzed with the help of hierarchical ontologies, such as the Gene Ontology (GO). We have developed a novel framework-based on the well established foundations of information theory - that allows for the evaluation of new types of hypotheses. The framework, encapsulated in Open Biomedical Ontology-Based Exploration and Search (OBOES), has already been applied in the investigation of different kinds of questions. The resulting framework enables the new field of information theoretic ontology-based analysis. We have applied this framework to create methods to re-engineer ontologies, explore fundamental questions on the evolution of biological complexity, determine optimal ontology terms for bioinformatics analysis, and quantify the usefulness of biofluids as proxies for tissues/diseases. In each case, we found that our methods provide novel, significant findings. An open source Java implementation of OBOES is available at: http://***. net.
We propose an infotainment presentation system that relies on eyegaze as an intuitive and unobtrusive input modality. The system analyzes eye movements in real-time to infer users' attention, visual interest, and ...
详细信息
ISBN:
(纸本)1595936408
We propose an infotainment presentation system that relies on eyegaze as an intuitive and unobtrusive input modality. The system analyzes eye movements in real-time to infer users' attention, visual interest, and preference regarding interface objects. The application consists of a virtual showroom where a team of two highly realistic 3D agents presents product items in an entertaining and attractive way. The presentation flow adapts to the user's attentiveness and interest, or lack thereof, and thus provides a more personalized and user-attentive experience of the presentation. Copyright 2007 ACM.
The First International Workshop on Database Preservation (PresDB' 07) organized by the UK Digital Curation Center (DCC) was held at the National e-science Center's e-scienceinstitute in Edinburgh during Marc...
详细信息
The First International Workshop on Database Preservation (PresDB' 07) organized by the UK Digital Curation Center (DCC) was held at the National e-science Center's e-scienceinstitute in Edinburgh during March 23, 2007 to identify new technical, economic, and legal issues arising due to shortcomings in database preservation. The scientific, scholarly, and cultural data sets are valuable and many of them are either impossible to reproduce or only be recovered at an enormous costs. Giorgos Flouris from Pisa, Italy has described a logic-based perspective on information preservation that allowed a formal definition of the data preservation problem and characterization of desirable properties of existing and future preservation methods. Peter Buneman from the school of informatics and Digital Curation Center of Edinburgh in UK claimed that DBMS history support is mainly designed for correction/recovery and flashback queries.
We study the Simplified Partial Digest Problem (SPDP), which is a mathematical model for a new simplified partial digest method of genome mapping. This method is easy for laboratory implementation and robust with resp...
详细信息
This paper concerns the problem of on-line (real-time) computation of solutions to the optimal switching time problem in hybrid systems. The systems under consideration are autonomous, and the performance measure to b...
详细信息
ISBN:
(纸本)9781424414970;1424414970
This paper concerns the problem of on-line (real-time) computation of solutions to the optimal switching time problem in hybrid systems. The systems under consideration are autonomous, and the performance measure to be optimized has the form of a cost functional defined on the state trajectory. The state variable cannot, however, be measured directly and it has to be estimated by a suitable observer. In this paper, we propose an on-line optimization algorithm based on the state observer, and derive bounds on its convergence rate.
In bioinformatics it is often desirable to combine data from various measurement sources and thus structured feature vectors are to be analyzed that possess different intrinsic blocking characteristics (e.g., differen...
ISBN:
(纸本)9781605603520
In bioinformatics it is often desirable to combine data from various measurement sources and thus structured feature vectors are to be analyzed that possess different intrinsic blocking characteristics (e.g., different patterns of missing values, observation noise levels, effective intrinsic dimensionalities). We propose a new machine learning tool, heterogeneous component analysis (HCA), for feature extraction in order to better understand the factors that underlie such complex structured heterogeneous data. HCA is a linear block-wise sparse Bayesian PCA based not only on a probabilistic model with block-wise residual variance terms but also on a Bayesian treatment of a block-wise sparse factor-loading matrix. We study various algorithms that implement our HCA concept extracting sparse heterogeneous structure by obtaining common components for the blocks and specific components within each block. Simulations on toy and bioinformatics data underline the usefulness of the proposed structured matrix factorization concept.
Multi-issue negotiations are a central component of many important coordination challenges. Almost all previous work in this area has assumed that negotiation issues are independent, making it relatively easy to find ...
详细信息
ISBN:
(纸本)9780769530277
Multi-issue negotiations are a central component of many important coordination challenges. Almost all previous work in this area has assumed that negotiation issues are independent, making it relatively easy to find high-quality agreements. In many real-world problem domains, however, issues are interdependent, making hard to find good agreement due to the nonlinearity of the agent's utility functions. The key challenge, in this context, is finding high-quality agreements without making unrealistic demands concerning how much agents reveal about their utilities. In this paper, we propose a protocol wherein the negotiating agents, working with the mediator, progress through a multi-phase narrowing of the space of possible agreements. We show that our method outperforms existing methods in large nonlinear utility spaces.
暂无评论