The main challenge in the area of reinforcement learning is scaling up to larger and more complex problems. Aiming at the scaling problem of reinforcement learning, a scalable reinforcement learning method, DCS-SRL, is proposed on the basis of divide-and-conquer strategy, and its convergence is proved. In this method, the learning problem in large state space or continuous state space is decomposed into multiple smaller subproblems. Given a specific learning algorithm, each subproblem can be solved independently with limited available resources. In the end, component solutions can be recombined to obtain the desired result. To address the question of prioritizing subproblems in the scheduler, a weighted priority scheduling algorithm is proposed. This scheduling algorithm ensures that computation is focused on regions of the problem space which are expected to be maximally productive. To expedite the learning process, a new parallel method, called DCS-SPRL, is derived from combining DCS-SRL with a parallel scheduling architecture. In the DCS-SPRL method, the subproblems will be distributed among processors that have the capacity to work in parallel. The experimental results show that learning based on DCS-SPRL has fast convergence speed and good scalability.

Listwise approaches are an important class of learning to rank, which utilizes automatic learning techniques to discover useful information. Most previous research on listwise approaches has focused on optimizing ranking models using weights and has used imprecisely labeled training data; optimizing ranking models using features was largely ignored thus the continuous performance improvement of these approaches was hindered. To address the limitations of previous listwise work, we propose a quasi-KNN model to discover the ranking of features and employ rank addition rule to calculate the weight of combination. On the basis of this, we propose three listwise algorithms, FeatureRank, BLFeatureRank, and DiffRank. The experimental results show that our proposed algorithms can be applied to a strict ordered ranking training set and gain better performance than state-of-the-art listwise algorithms.

Increasingly powerful computational technology has caused enormous data growth both in size and complexity. A key issue is how to organize the data to adapt the challenges of data analysis. This paper borrows ideas from the Internet of things (IOT) into the digital world and organize the data entities to form a network, the Internet of data (IOD), which has huge potential in data-intensive applications. In the IOD, data hiding technology is utilized to embed virtual tags, which record all the activities of the data entities since they are created, into every data entity in the system. The IOD aims to organize the data to be interconnected as a network and collect useful information for data identification, data tracing, data vitalization and further data analysis.

Powerful storage, high performance and scalability are the most important issues for analytical databases. These three factors interact with each other, for example, powerful storage needs less scalability but higher performance, high performance means less consumption of indexes and other materializations for storage and fewer processing nodes, larger scale relieves stress on powerful storage and the high performance processing engine. Some analytical databases (ParAccel, Teradata) bind their performance with advanced hardware supports, some (Asterdata, Greenplum) rely on the high scalability framework of MapReduce, some (MonetDB, Sybase IQ, Vertica) highlight performance on processing engine and storage engine. All these approaches can be integrated into an storage-performance-scalability (SP- S) model, and future large scale analytical processing can be built on moderate clusters to minimize expensive hardware dependency. The most important thing is a simple software framework is fundamental to maintain pace with the development of hardware technologies. In this paper, we propose a schema-aware on-line analytical processing (OLAP) model with deep optimization from native features of the star or snowflake schema. The OLAP model divides the whole process into several stages, each stage pipes its output to the next stage, we minimize the size of output data in each stage, whether in central processing or clustered processing. We extend this mechanism to cluster processing using two major techniques, one is using NetMemory as a broadcasting protocol based dimension mirror synchronizing buffer, the other is Received June 24, 2011; accepted August 16, 2012 E-mail: shingle@ruc.edu.cn predicate-vector based DDTA-OLAP cluster model which can minimize the data dependency of star-join using bitmap vectors. Our OLAP model aims to minimize network transmission cost (MiNT in short) for OLAP clusters and support a scalable but simple distributed storagemodel for large scale clustering processing. Finally, the experimental results show the speedup and scalability performance.

To manage dynamic access control and deter pirate attacks on outsourced databases, a dynamic access control scheme with tracing is proposed. In our scheme, we introduce the traitor tracing idea into outsource databases, and employ a polynomial function and filter function as the basic means of constructing encryption and decryption procedures to reduce computation, communication, and storage overheads. Compared to previous access control schemes for outsourced databases, our scheme can not only protect sensitive data from leaking and perform scalable encryption at the server side without shipping the outsourced data back to the data owner when group membership is changed, but also provide trace-and-revoke features.When malicious users clone and sell their decryption keys for profit, our scheme can trace the decryption keys to the malicious users and revoke them. Furthermore, our scheme avoids massive message exchanges for establishing the decryption key between the data owner and the user. Compared to previously proposed publickey traitor tracing schemes, our scheme can simultaneously achieve full collusion resistance, full recoverability, full revocation, and black-box traceability. The proof of security and analysis of performance show that our scheme is secure and efficient.

Set queries are an important topic and have attracted a lot of attention. Earlier research mainly concentrated on set containment queries. In this paper we focus on the T-Overlap query which is the foundation of the set similarity query. To address this issue, unlike traditional algorithms that are based on an inverted index, we design a new paradigm based on the prefix tree (trie) called the expanded trie index (ETI) which expands the trie node structure by adding some new properties. Based on ETI, we convert the TOverlap problem to finding query nodes with specific query depth equaling to T and propose a new algorithm called TSimilarity to solve T-Overlap efficiently. Then we carry out a three-step framework to extend T-Overlap to other similarity predicates. Extensive experiments are carried out to compare T-Similarity with other inverted index based algorithms from cardinality of query, overlap threshold, dataset size, the number of distinct elements and so on. Results show that T-Similarity outperforms the state-of-the-art algorithms in many aspects.

A vehicular ad-hoc network (VANET) can be visualized as a network of moving vehicles communicating in an asynchronous and autonomous fashion. Efficient and scalable information dissemination in VANET applications is a major challenge due to the movement of vehicles which causes unpredictable changes in network topology. The publish/ subscribe communication paradigm provides decoupling in time, space, and synchronization between communicating entities, and presents itself as an elegant solution for information dissemination for VANET like environments. In this paper, we propose our approach for information dissemination which utilizes publish/subscribe and distributed hash table (DHT) based overlay networks. In our approach, we assume a hybrid VANET consisting of stationary info-stations and moving vehicles. These info-stations are installed at every major intersection of the city and vehicles can take the role of publisher, subscriber, or broker depending upon the context. The info-stations form a DHT based broker overlay among themselves and act as rendezvous points for related publications and subscriptions. Further, info-stations also assist in locating vehicles that have subscribed to information items. We consider different possible deployments of this hybrid VANET with respect to the number of info-stations and their physical connectivity with each other. We perform Received August 22, 2011; accepted January 21, 2012 E-mail: tulika_mishra@rediffmail.com simulations to assess the performance of our approach in these different deployment scenarios and discuss their applicability in urban and semi-urban areas.

When reengineering a monolithic application to be a distributed one, programmers always have to decide how many distributed parts the application should be partitioned, and write many codes related to where a partwill be placed on network nodes and how these parts communicate with each other through the network. These codes usually have nothing to do with the business functions of the application, and they are laborious to write. In addition, as the distribution architecture of the application is finalized beforehand, it may not adapt well to the ever-changing execution environment. In this paper, we propose DPartner, an automatic partitioning system, to help programmers create a distributed Java application without explicitly writing the distribution-related codes. Unlike the other partitioning systems, DPartner does not partition an application directly into the coarse-grained client and server. Instead, it first partitions the application into several modules where each module performs a relatively independent business function of the application. Then it makes these modules be distributable through automatic bytecode rewriting. These modules can distribute in different nodes and cooperate to work just as the original monolithic application. Such a module-based partitioning approach enables a relatively easy reshaping of the distribution architecture of an application, which facilitates the application adapt to the environmental changes without manual recoding or repartitioning with regard to distribution. This paper gives the detailed design of DPartner, and evaluates it using real-world applications. The evaluation results demonstrate the effectiveness and efficiency of DPartner.

In this work, we propose several new methods for detecting photographic composites using circles. In particular, we focus on three kinds of scenes: (1) two coplanar circles with the same radius; (2) a single circle with its discriminable center; (3) a single circle with geometric constraints for camera calibration. For two circles’ situation, we first estimate the focal length based on the equality of the sizes of two coplanar circles, and then estimate the normal vector of the world circle plane. Inconsistencies in the angles among the normal vectors (Each circle determines a normal vector) are used as evidence of tampering. On the other hand, for the single circle case, we warp the circle to make metric measurement. To demonstrate the effectiveness of the approach, we present results for synthetic and visually plausible composite images.

Creating realistic 3D tree models in a convenient way is a challenge in game design and movie making due to diversification and occlusion of tree structures. Current sketch-based and image-based approaches for fast tree modeling have limitations in effect and speed, and they generally include complex parameter adjustment, which brings difficulties to novices. In this paper, we present a simple method for quickly generating various 3D tree models from freehand sketches without parameter adjustment. On two input images, the user draws strokes representing the main branches and crown silhouettes of a tree. The system automatically produces a 3D tree at high speed. First, two 2D skeletons are built from strokes, and a 3D tree structure resembling the input sketches is built by branch retrieval from the 2D skeletons. Small branches are generated within the sketched 2D crown silhouettes based on self-similarity and angle restriction. This system is demonstrated on a variety of examples. It maintains the main features of a tree: the main branch structure and crown shape, and can be used as a convenient tool for tree simulation and design.

In recent years, geometry-based image and video processing methods have aroused significant interest. This paper considers progress from four aspects: geometric characteristics and shape, geometric transformations, embedded geometric structure, and differential geometry methods. Current research trends are also pointed out.