Generalized pseudo-likelihood estimates for markov random fields on lattice

In this paper we generalize Besag's pseudo-likelihood function for spatial statistical models on a region of a lattice. The correspondingly defined maximum generalized pseudo-likelihood estimates (MGPLEs) are natural extensions of Besag's maximum pseudo-likelihood estimate (MPLE). The MGPLEs connect the MPLE and the maximum likelihood estimate. We carry out experimental calculations of the MGPLEs for spatial processes on the lattice. These simulation results clearly show better performances of the MGPLEs than the MPLE, and the performances of differently defined MGPLEs are compared. These are also illustrated by the application to two real data sets.

Hierarchical semi-markov conditional random fields for recursive sequential data

Inspired by the hierarchical hidden Markov models (HHMM), we present the hierarchical semi-Markov conditional random field (HSCRF), a generalisation of embedded undirected Markov chains to model complex hierarchical, nested Markov processes. It is parameterised in a discriminative framework and has polynomial time algorithms for learning and inference. Importantly, we develop efficient algorithms for learning and constrained inference in a partially-supervised setting, which is important issue in practice where labels can only be obtained sparsely. We demonstrate the HSCRF in two applications: (i) recognising human activities of daily living (ADLs) from indoor surveillance cameras, and (ii) noun-phrase chunking. We show that the HSCRF is capable of learning rich hierarchical models with reasonable accuracy in both fully and partially observed data cases.

A hybrid cuckoo search and variable neighborhood descent for single and multiobjective scheduling problems

Cuckoo search (CS) is a relatively new meta-heuristic that has proven its strength in solving continuous optimization problems. This papers applies cuckoo search to the class of sequencing problems by hybridizing it with a variable neighborhood descent local search for enhancing the quality of the obtained solutions. The Lévy flight operator proposed in the original CS is modified to address the discrete nature of scheduling problems. Two well-known problems are used to demonstrate the effectiveness of the proposed hybrid CS approach. The first is the NP-hard single objective problem of minimizing the weighted total tardiness time (Formula presented.) and the second is the multiobjective problem of minimizing the flowtime ¯ and the maximum tardiness Tmaxfor single machine (Formula presented.). For the first problem, computational results show that the hybrid CS is able to find the optimal solutions for all benchmark test instances with 40, 50, and 100 jobs and for most instances with 150, 200, 250, and 300 jobs. For the second problem, the hybrid CS generated solutions on and very close to the exact Pareto fronts of test instances with 10, 20, 30, and 40 jobs. In general, the results reveal that the hybrid CS is an adequate and robust method for tackling single and multiobjective scheduling problems.

Tree-based iterated local search for Markov random fields with applications in image analysis

The maximum a posteriori assignment for general structure Markov random fields is computationally intractable. In this paper, we exploit tree-based methods to efficiently address this problem. Our novel method, named Tree-based Iterated Local Search (T-ILS), takes advantage of the tractability of tree-structures embedded within MRFs to derive strong local search in an ILS framework. The method efficiently explores exponentially large neighborhoods using a limited memory without any requirement on the cost functions. We evaluate the T-ILS on a simulated Ising model and two real-world vision problems: stereo matching and image denoising. Experimental results demonstrate that our methods are competitive against state-of-the-art rivals with significant computational gain.

Ordinal random fields for recommender systems

Recommender Systems heavily rely on numerical preferences, whereas the importance of ordinal preferences has only been recognised in recent works of Ordinal Matrix Factorisation (OMF). Although the OMF can effectively exploit ordinal properties, it captures only the higher-order interactions among users and items, without considering the localised interactions properly. This paper employs Markov Random Fields (MRF) to investigate the localised interactions, and proposes a unified model called Ordinal Random Fields (ORF) to take advantages of both the representational power of the MRF and the ease of modelling ordinal preferences by the OMF. Experimental result on public datasets demonstrates that the proposed ORF model can capture both types of interactions, resulting in improved recommendation accuracy.

Preference relation-based markov random fields

A preference relation-based Top-N recommendation approach, PrefMRF, is proposed to capture both the second-order and the higher-order interactions among users and items. Traditionally Top-N recommendation was achieved by predicting the item ratings fi rst, and then inferring the item rankings, based on the assumption of availability of explicit feed-backs such as ratings, and the assumption that optimizing the ratings is equivalent to optimizing the item rankings. Nevertheless, both assumptions are not always true in real world applications. The proposed PrefMRF approach drops these assumptions by explicitly exploiting the preference relations, a more practical user feedback. Comparing to related work, the proposed PrefMRF approach has the unique property of modeling both the second-order and the higher-order interactions among users and items. To the best of our knowledge, this is the first time both types of interactions have been captured in preference relation-based method. Experiment results on public datasets demonstrate that both types of interactions have been properly captured, and signifi cantly improved Top-N recommendation performance has been achieved.

Collaborative filtering via sparse Markov random fields

Recommender systems play a central role in providing individualized access to information and services. This paper focuses on collaborative filtering, an approach that exploits the shared structure among mind-liked users and similar items. In particular, we focus on a formal probabilistic framework known as Markov random fields (MRF). We address the open problem of structure learning and introduce a sparsity-inducing algorithm to automatically estimate the interaction structures between users and between items. Item-item and user-user correlation networks are obtained as a by-product. Large-scale experiments on movie recommendation and date matching datasets demonstrate the power of the proposed method.

Preference relation-based Markov Random Fields for recommender systems

A preference relation-based Top-N recommendation approach is proposed to capture both second-order and higher-order interactions among users and items. Traditionally Top-N recommendation was achieved by predicting the item ratings first, and then inferring the item rankings, based on the assumption of availability of explicit feedback such as ratings, and the assumption that optimizing the ratings is equivalent to optimizing the item rankings. Nevertheless, both assumptions are not always true in real world applications. The proposed approach drops these assumptions by exploiting preference relations, a more practical user feedback. Furthermore, the proposed approach enjoys the representational power of Markov Random Fields thus side information such as item and user attributes can be easily incorporated. Comparing to related work, the proposed approach has the unique property of modeling both second-order and higher-order interactions among users and items. To the best of our knowledge, this is the first time both types of interactions have been captured in preference-relation based methods. Experimental results on public datasets demonstrate that both types of interactions have been properly captured, and significantly improved Top-N recommendation performance has been achieved.

AdaBoost.MRF: boosted Markov random forests and application to multilevel activity recognition

Activity recognition is an important issue in building intelligent monitoring systems. We address the recognition of multilevel activities in this paper via a conditional Markov random field (MRF), known as the dynamic conditional random field (DCRF). Parameter estimation in general MRFs using maximum likelihood is known to be computationally challenging (except for extreme cases), and thus we propose an efficient boosting-based algorithm AdaBoost.MRF for this task. Distinct from most existing work, our algorithm can handle hidden variables (missing labels) and is particularly attractive for smarthouse domains where reliable labels are often sparsely observed. Furthermore, our method works exclusively on trees and thus is guaranteed to converge. We apply the AdaBoost.MRF algorithm to a home video surveillance application and demonstrate its efficacy.

Learning discriminative sequence models from partially labelled data for activity recognition

Recognising daily activity patterns of people from low-level sensory data is an important problem. Traditional approaches typically rely on generative models such as the hidden Markov models and training on fully labelled data. While activity data can be readily acquired from pervasive sensors, e.g. in smart environments, providing manual labels to support fully supervised learning is often expensive. In this paper, we propose a new approach based on partially-supervised training of discriminative sequence models such as the conditional random field (CRF) and the maximum entropy Markov model (MEMM). We show that the approach can reduce labelling effort, and at the same time, provides us with the flexibility and accuracy of the discriminative framework. Our experimental results in the video surveillance domain illustrate that these models can perform better than their generative counterpart (i.e. the partially hidden Markov model), even when a substantial amount of labels are unavailable.

Locally sparsified compressive sensing in magnetic resonance imaging

Magnetic Resonance Imaging (MRI) is a widely used technique for acquiring images of human organs/tissues. Due to its complex imaging process, it consumes a lot of time to produce a high quality image. Compressive Sensing (CS) has been used by researchers for rapid MRI. It uses a global sparsity constraint with variable density random sampling and L1 minimisation. This work intends to speed up the imaging process by exploiting the non-uniform sparsity in the MR images. Locally Sparsified CS suggests that the image can be even better sparsified by applying local sparsity constraints. The image produced by local CS can further reduce the sample set. This paper establishes the basis for a methodology to exploit non-uniform nature of sparsity and to make the MRI process time efficient by using local sparsity constraints.

Leveraging visual features and hierarchical dependencies for conference information extraction

Traditional information extraction methods mainly rely on visual feature assisted techniques; but without considering the hierarchical dependencies within the paragraph structure, some important information is missing. This paper proposes an integrated approach for extracting academic information from conference Web pages. Firstly, Web pages are segmented into text blocks by applying a new hybrid page segmentation algorithm which combines visual feature and DOM structure together. Then, these text blocks are labeled by a Tree-structured Random Fields model, and the block functions are differentiated using various features such as visual features, semantic features and hierarchical dependencies. Finally, an additional post-processing is introduced to tune the initial annotation results. Our experimental results on real-world data sets demonstrated that the proposed method is able to effectively and accurately extract the needed academic information from conference Web pages. © 2013 Springer-Verlag.

How do perceptions of local neighborhood relate to adolescents` walking and cycling?

Purpose. To examine how perceptions of the local neighborhood relate to adolescents' walking and cycling. Design. Exploratory cross-sectional study. Setting. Birth cohort from the Nepean Hospital, Sydney, Australia. Subjects. Three hundred forty-seven adolescents (79.1 % response rate; 49.6% boys; mean age - 13.0 ± 0.2 years) and their parents. Measures. Self-report and parental-report questionnaires. Results. Multiple linear regressions, adjusted for level of maternal education, revealed that boys who reported having many peers to hang out with locally, cycled for recreation (β = 0.242, p = .006) or for transport (β = 0.141, p = . 046) more often, and walked for transport for longer (β = 0.129, p = .024) on weekdays. For girls this variable was related to cycling for recreation on weekends (β = 0.164, p = .006) and walking to school (β = 0.118, p = .002). Adolescents who waved/talked to neighbors walked for transport more often (boys, β = 0.149, p = .037; girls, β = 0.119, p = .012). Girls who perceived local roads to be safe spent more time walking for transport on weekdays (β = 0.183, p = .007) and for exercise on weekends (β = 0.184, p = . 034). Parents' perception of heavy traffic was negatively associated with boys' walking for transport (β = -0.138, p = .037) and many aspects of girls' walking and cycling. Conclusion. Social interaction and road safety may be important predictors of adolescents' walking and cycling in their neighborhood. Limitations are the use of self-report and cross-sectional data. Longitudinal studies may clarify these relations.

Automotive can network response time analysis with variable jitter

Many methods to calculate message latencies for Controller Area Network (CAN) have previously been presented based upon the static worst-case behaviour of the system. With the use of modern simulation tools however, the behaviour of CAN networks can be simulated dynamically in order to find the likely worst-case response times for CAN messages. This paper shows the development of an automotive body control network model to be used as the basis for further simulations. A method to simulate the Worst-Case Response Time of this model is then presented, taking into account random queuing jitter.