'Explicit Learning: an Effort towards Human Scheduling Algorithms'

Abstract Scheduling problems are generally NP-hard combinatorial problems, and a lot of research has been done to solve these problems heuristically. However, most of the previous approaches are problem-specific and research into the development of a general scheduling algorithm is still in its infancy. Mimicking the natural evolutionary process of the survival of the fittest, Genetic Algorithms (GAs) have attracted much attention in solving difficult scheduling problems in recent years. Some obstacles exist when using GAs: there is no canonical mechanism to deal with constraints, which are commonly met in most real-world scheduling problems, and small changes to a solution are difficult. To overcome both difficulties, indirect approaches have been presented (in [1] and [2]) for nurse scheduling and driver scheduling, where GAs are used by mapping the solution space, and separate decoding routines then build solutions to the original problem. In our previous indirect GAs, learning is implicit and is restricted to the efficient adjustment of weights for a set of rules that are used to construct schedules. The major limitation of those approaches is that they learn in a non-human way: like most existing construction algorithms, once the best weight combination is found, the rules used in the construction process are fixed at each iteration. However, normally a long sequence of moves is needed to construct a schedule and using fixed rules at each move is thus unreasonable and not coherent with human learning processes. When a human scheduler is working, he normally builds a schedule step by step following a set of rules. After much practice, the scheduler gradually masters the knowledge of which solution parts go well with others. He can identify good parts and is aware of the solution quality even if the scheduling process is not completed yet, thus having the ability to finish a schedule by using flexible, rather than fixed, rules. In this research we intend to design more human-like scheduling algorithms, by using ideas derived from Bayesian Optimization Algorithms (BOA) and Learning Classifier Systems (LCS) to implement explicit learning from past solutions. BOA can be applied to learn to identify good partial solutions and to complete them by building a Bayesian network of the joint distribution of solutions [3]. A Bayesian network is a directed acyclic graph with each node corresponding to one variable, and each variable corresponding to individual rule by which a schedule will be constructed step by step. The conditional probabilities are computed according to an initial set of promising solutions. Subsequently, each new instance for each node is generated by using the corresponding conditional probabilities, until values for all nodes have been generated. Another set of rule strings will be generated in this way, some of which will replace previous strings based on fitness selection. If stopping conditions are not met, the Bayesian network is updated again using the current set of good rule strings. The algorithm thereby tries to explicitly identify and mix promising building blocks. It should be noted that for most scheduling problems the structure of the network model is known and all the variables are fully observed. In this case, the goal of learning is to find the rule values that maximize the likelihood of the training data. Thus learning can amount to 'counting' in the case of multinomial distributions. In the LCS approach, each rule has its strength showing its current usefulness in the system, and this strength is constantly assessed [4]. To implement sophisticated learning based on previous solutions, an improved LCS-based algorithm is designed, which consists of the following three steps. The initialization step is to assign each rule at each stage a constant initial strength. Then rules are selected by using the Roulette Wheel strategy. The next step is to reinforce the strengths of the rules used in the previous solution, keeping the strength of unused rules unchanged. The selection step is to select fitter rules for the next generation. It is envisaged that the LCS part of the algorithm will be used as a hill climber to the BOA algorithm. This is exciting and ambitious research, which might provide the stepping-stone for a new class of scheduling algorithms. Data sets from nurse scheduling and mall problems will be used as test-beds. It is envisaged that once the concept has been proven successful, it will be implemented into general scheduling algorithms. It is also hoped that this research will give some preliminary answers about how to include human-like learning into scheduling algorithms and may therefore be of interest to researchers and practitioners in areas of scheduling and evolutionary computation. References 1. Aickelin, U. and Dowsland, K. (2003) 'Indirect Genetic Algorithm for a Nurse Scheduling Problem', Computer & Operational Research (in print). 2. Li, J. and Kwan, R.S.K. (2003), 'Fuzzy Genetic Algorithm for Driver Scheduling', European Journal of Operational Research 147(2): 334-344. 3. Pelikan, M., Goldberg, D. and Cantu-Paz, E. (1999) 'BOA: The Bayesian Optimization Algorithm', IlliGAL Report No 99003, University of Illinois. 4. Wilson, S. (1994) 'ZCS: A Zeroth-level Classifier System', Evolutionary Computation 2(1), pp 1-18.

'Explicit Learning: an Effort towards Human Scheduling Algorithms'

Abstract Scheduling problems are generally NP-hard combinatorial problems, and a lot of research has been done to solve these problems heuristically. However, most of the previous approaches are problem-specific and research into the development of a general scheduling algorithm is still in its infancy. Mimicking the natural evolutionary process of the survival of the fittest, Genetic Algorithms (GAs) have attracted much attention in solving difficult scheduling problems in recent years. Some obstacles exist when using GAs: there is no canonical mechanism to deal with constraints, which are commonly met in most real-world scheduling problems, and small changes to a solution are difficult. To overcome both difficulties, indirect approaches have been presented (in [1] and [2]) for nurse scheduling and driver scheduling, where GAs are used by mapping the solution space, and separate decoding routines then build solutions to the original problem. In our previous indirect GAs, learning is implicit and is restricted to the efficient adjustment of weights for a set of rules that are used to construct schedules. The major limitation of those approaches is that they learn in a non-human way: like most existing construction algorithms, once the best weight combination is found, the rules used in the construction process are fixed at each iteration. However, normally a long sequence of moves is needed to construct a schedule and using fixed rules at each move is thus unreasonable and not coherent with human learning processes. When a human scheduler is working, he normally builds a schedule step by step following a set of rules. After much practice, the scheduler gradually masters the knowledge of which solution parts go well with others. He can identify good parts and is aware of the solution quality even if the scheduling process is not completed yet, thus having the ability to finish a schedule by using flexible, rather than fixed, rules. In this research we intend to design more human-like scheduling algorithms, by using ideas derived from Bayesian Optimization Algorithms (BOA) and Learning Classifier Systems (LCS) to implement explicit learning from past solutions. BOA can be applied to learn to identify good partial solutions and to complete them by building a Bayesian network of the joint distribution of solutions [3]. A Bayesian network is a directed acyclic graph with each node corresponding to one variable, and each variable corresponding to individual rule by which a schedule will be constructed step by step. The conditional probabilities are computed according to an initial set of promising solutions. Subsequently, each new instance for each node is generated by using the corresponding conditional probabilities, until values for all nodes have been generated. Another set of rule strings will be generated in this way, some of which will replace previous strings based on fitness selection. If stopping conditions are not met, the Bayesian network is updated again using the current set of good rule strings. The algorithm thereby tries to explicitly identify and mix promising building blocks. It should be noted that for most scheduling problems the structure of the network model is known and all the variables are fully observed. In this case, the goal of learning is to find the rule values that maximize the likelihood of the training data. Thus learning can amount to 'counting' in the case of multinomial distributions. In the LCS approach, each rule has its strength showing its current usefulness in the system, and this strength is constantly assessed [4]. To implement sophisticated learning based on previous solutions, an improved LCS-based algorithm is designed, which consists of the following three steps. The initialization step is to assign each rule at each stage a constant initial strength. Then rules are selected by using the Roulette Wheel strategy. The next step is to reinforce the strengths of the rules used in the previous solution, keeping the strength of unused rules unchanged. The selection step is to select fitter rules for the next generation. It is envisaged that the LCS part of the algorithm will be used as a hill climber to the BOA algorithm. This is exciting and ambitious research, which might provide the stepping-stone for a new class of scheduling algorithms. Data sets from nurse scheduling and mall problems will be used as test-beds. It is envisaged that once the concept has been proven successful, it will be implemented into general scheduling algorithms. It is also hoped that this research will give some preliminary answers about how to include human-like learning into scheduling algorithms and may therefore be of interest to researchers and practitioners in areas of scheduling and evolutionary computation. References 1. Aickelin, U. and Dowsland, K. (2003) 'Indirect Genetic Algorithm for a Nurse Scheduling Problem', Computer & Operational Research (in print). 2. Li, J. and Kwan, R.S.K. (2003), 'Fuzzy Genetic Algorithm for Driver Scheduling', European Journal of Operational Research 147(2): 334-344. 3. Pelikan, M., Goldberg, D. and Cantu-Paz, E. (1999) 'BOA: The Bayesian Optimization Algorithm', IlliGAL Report No 99003, University of Illinois. 4. Wilson, S. (1994) 'ZCS: A Zeroth-level Classifier System', Evolutionary Computation 2(1), pp 1-18.

Iron Transformation Pathways and Redox Micro-Environments in Seafloor Sulfide-Mineral Deposits: Spatially Resolved Fe XAS and delta Fe-57/54 Observations

Hydrothermal sulfide chimneys located along the global system of oceanic spreading centers are habitats for microbial life during active venting. Hydrothermally extinct, or inactive, sulfide deposits also host microbial communities at globally distributed sites. The main goal of this study is to describe Fe transformation pathways, through precipitation and oxidation-reduction (redox) reactions, and examine transformation products for signatures of biological activity using Fe mineralogy and stable isotope approaches. The study includes active and inactive sulfides from the East Pacific Rise 9 degrees 50'N vent field. First, the mineralogy of Fe(III)-bearing precipitates is investigated using microprobe X-ray absorption spectroscopy (RXAS) and X-ray diffraction (mu XRD). Second, laser-ablation (LA) and micro-drilling (MD) are used to obtain spatially-resolved Fe stable isotope analysis by multicollector-inductively coupled plasma-mass spectrometry (MC-ICP-MS). Eight Fe -bearing minerals representing three mineralogical classes are present in the samples: oxyhydroxides, secondary phyllosilicates, and sulfides. For Fe oxyhydroxides within chimney walls and layers of Si-rich material, enrichments in both heavy and light Fe isotopes relative to pyrite are observed, yielding a range of delta Fe-57 values up to 6 parts per thousand. Overall, several pathways for Fe transformation are observed. Pathway 1 is characterized by precipitation of primary sulfide minerals from Fe(II)aq-rich fluids in zones of mixing between vent fluids and seawater. Pathway 2 is also consistent with zones of mixing but involves precipitation of sulfide minerals from Fe(II)aq generated by Fe(III) reduction. Pathway 3 is direct oxidation of Fe(II) aq from hydrothermal fluids to form Fe(III) precipitates. Finally, Pathway 4 involves oxidative alteration of pre-existing sulfide minerals to form Fe(III). The Fe mineralogy and isotope data do not support or refute a unique biological role in sulfide alteration. The findings reveal a dynamic range of Fe transformation pathways consistent with a continuum of micro-environments having variable redox conditions. These micro-environments likely support redox cycling of Fe and S and are consistent with culture-dependent and -independent assessments of microbial physiology and genetic diversity of hydrothermal sulfide deposits.

The book of choice ferns for the garden, conservatory, and stove. Describing and giving explicit cultural directions for the best and most striking ferns and selaginellas in cultivation. Illustrated with coloured plates and numerous wood engravings, specially prepared for this work.

vol.I. Introduction to Athyrium.--vol.II. Blechnum to Nothochlaena.--vol.III. Ochropteris to Woodwardia, and Selaginella.

Mode locking in a spatially extended neuron model: active soma and compartmental tree

Understanding the mode-locked response of excitable systems to periodic forcing has important applications in neuroscience. For example it is known that spatially extended place cells in the hippocampus are driven by the theta rhythm to generate a code conveying information about spatial location. Thus it is important to explore the role of neuronal dendrites in generating the response to periodic current injection. In this paper we pursue this using a compartmental model, with linear dynamics for each compartment, coupled to an active soma model that generates action potentials. By working with the piece-wise linear McKean model for the soma we show how the response of the whole neuron model (soma and dendrites) can be written in closed form. We exploit this to construct a stroboscopic map describing the response of the spatially extended model to periodic forcing. A linear stability analysis of this map, together with a careful treatment of the non-differentiability of the soma model, allows us to construct the Arnol'd tongue structure for 1:q states (one action potential for q cycles of forcing). Importantly we show how the presence of quasi-active membrane in the dendrites can influence the shape of tongues. Direct numerical simulations confirm our theory and further indicate that resonant dendritic membrane can enlarge the windows in parameter space for chaotic behavior. These simulations also show that the spatially extended neuron model responds differently to global as opposed to point forcing. In the former case spatio-temporal patterns of activity within an Arnol'd tongue are standing waves, whilst in the latter they are traveling waves.

Migration, prospecting, dispersal? What host movement matters for infectious agent circulation?

Spatial disease ecology is emerging as a new field that requires the integration of complementary approaches to address how the distribution and movements of hosts and parasites may condition the dynamics of their interactions. In this context, migration, the seasonal movement of animals to different zones of their distribution, is assumed to play a key role in the broad scale circulation of parasites and pathogens. Nevertheless, migration is not the only type of host movement that can influence the spatial ecology, evolution, and epidemiology of infectious diseases. Dispersal, the movement of individuals between the location where they were born or bred to a location where they breed, has attracted attention as another important type of movement for the spatial dynamics of infectious diseases. Host dispersal has notably been identified as a key factor for the evolution of host-parasite interactions as it implies gene flow among local host populations and thus can alter patterns of coevolution with infectious agents across spatial scales. However, not all movements between host populations lead to dispersal per se. One type of host movement that has been neglected, but that may also play a role in parasite spread is prospecting, i.e., movements targeted at selecting and securing new habitat for future breeding. Prospecting movements, which have been studied in detail in certain social species, could result in the dispersal of infectious agents among different host populations without necessarily involving host dispersal. In this article, we outline how these various types of host movements might influence the circulation of infectious disease agents and discuss methodological approaches that could be used to assess their importance. We specifically focus on examples from work on colonial seabirds, ticks, and tick-borne infectious agents. These are convenient biological models because they are strongly spatially structured and involve relatively simple communities of interacting species. Overall, this review emphasizes that explicit consideration of the behavioral and population ecology of hosts and parasites is required to disentangle the relative roles of different types of movement for the spread of infectious diseases.

EXPLICIT WRITTEN CORRECTIVE FEEDBACK AND LANGUAGE APTITUDE IN SLA: IMPLICATIONS FOR IMPROVEMENT OF LINGUISTIC ACCURACY

Most second language researchers agree that there is a role for corrective feedback in second language writing classes. However, many unanswered questions remain concerning which linguistic features to target and the type and amount of feedback to offer. This study examined two new pieces of writing by 151 learners of English as a Second Language (ESL), in order to investigate the effect of direct and metalinguistic written feedback on errors with the simple past tense, the present perfect tense, dropped pronouns, and pronominal duplication. This inquiry also considered the extent to which learner differences in language-analytic ability (LAA), as measured by the LLAMA F, mediated the effects of these two types of explicit written corrective feedback. Learners in the feedback groups were provided with corrective feedback on two essays, after which learners in all three groups completed two additional writing tasks to determine whether or not the provision of corrective feedback led to greater gains in accuracy compared to no feedback. Both treatment groups, direct and metalinguistic, performed better than the comparison group on new pieces of writing immediately following the treatment sessions, yet direct feedback was more durable than metalinguistic feedback for one structure, the simple past tense. Participants with greater LAA proved more likely to achieve gains in the direct feedback group than in the metalinguistic group, whereas learners with lower LAA benefited more from metalinguistic feedback. Overall, the findings of the present study confirm the results of prior studies that have found a positive role for written corrective feedback in instructed second language acquisition.

Explicit Reportorium perutile de pravitate hereticorum et apostatarum

Detalles físicos: Encuadernación flexible en pergamino, en regular estado de conservación (arrugado, manchado). Visibles maculaturas. Guardas elaboradas con sobrantes de imprenta. Texto sin encabezamiento, bien impreso en 38 líneas sobre papel de calidad con caracteres góticos, de dos tamaños. Signaturas. Colofón. Visibles manchas de humedad. Dotación según notas de la fichas en antiguo catalogo. Incunable. Título y datos de publicación tomados del colofón. Incluye índice. Desde "abiuratio" (= abjuración), hasta "zizania" (=cizaña). Es un manual, en orden alfabético, acerca de cómo proceder con los herejes y los apóstatas en diversos casos. Tanto en materias de derecho civil como de derecho eclesiástico.

Assessing the risk of the evolution of resistance to pesticides using spatially complex simulation models.

2008

Detailed analysis of a conservative difference approximation for the time fractional diffusion equation

Diffusion equations that use time fractional derivatives are attractive because they describe a wealth of problems involving non-Markovian Random walks. The time fractional diffusion equation (TFDE) is obtained from the standard diffusion equation by replacing the first-order time derivative with a fractional derivative of order α ∈ (0, 1). Developing numerical methods for solving fractional partial differential equations is a new research field and the theoretical analysis of the numerical methods associated with them is not fully developed. In this paper an explicit conservative difference approximation (ECDA) for TFDE is proposed. We give a detailed analysis for this ECDA and generate discrete models of random walk suitable for simulating random variables whose spatial probability density evolves in time according to this fractional diffusion equation. The stability and convergence of the ECDA for TFDE in a bounded domain are discussed. Finally, some numerical examples are presented to show the application of the present technique.

Capabilities : describing what services can do

The ability of agents and services to automatically locate and interact with unknown partners is a goal for both the semantic web and web services. This, \serendipitous interoperability", is hindered by the lack of an explicit means of describing what services (or agents) are able to do, that is, their capabilities. At present, informal descriptions of what services can do are found in \documentation" elements; or they are somehow encoded in operation names and signatures. We show, by ref- erence to existing service examples, how ambiguous and imprecise capa- bility descriptions hamper the attainment of automated interoperability goals in the open, global web environment. In this paper we propose a structured, machine readable description of capabilities, which may help to increase the recall and precision of service discovery mechanisms. Our capability description draws on previous work in capability and process modeling and allows the incorporation of external classi¯cation schemes. The capability description is presented as a conceptual meta model. The model supports conceptual queries and can be used as an extension to the DAML-S Service Pro¯le.

Extending conceptual models for Web based applications

The next phase envisioned for the World Wide Web is automated ad-hoc interaction between intelligent agents, web services, databases and semantic web enabled applications. Although at present this appears to be a distant objective, there are practical steps that can be taken to advance the vision. We propose an extension to classical conceptual models to allow the definition of application components in terms of public standards and explicit semantics, thus building into web-based applications, the foundation for shared understanding and interoperability. The use of external definitions and the need to store outsourced type information internally, brings to light the issue of object identity in a global environment, where object instances may be identified by multiple externally controlled identification schemes. We illustrate how traditional conceptual models may be augmented to recognise and deal with multiple identities.

The future of office property

The following paper considers the question, where to office property? In doing so, it focuses, in the first instance, on identifying and describing a selection of key forces for change present within the contemporary operating environment in which office property functions. Given the increasingly complex, dynamic and multi-faceted character of this environment, the paper seeks to identify only the primary forces for change, within the context of the future of office property. These core drivers of change have, for the purposes of this discussion, been characterised as including a range of economic, demographic and socio-cultural factors, together with developments in information and communication technology. Having established this foundation, the paper proceeds to consider the manner in which these forces may, in the future, be manifested within the office property market. Comment is offered regarding the potential future implications of these forces for change together with their likely influence on the nature and management of the physical asset itself. Whilst no explicit time horizon has been envisioned in the preparation of this paper particular attention has been accorded short to medium term trends, that is, those likely to emerge in the office property marketplace over the coming two decades. Further, the paper considers the question posed, in respect of the future of office property, in the context of developed western nations. The degree of commonality seen in these mature markets is such that generalisations may more appropriately and robustly be applied. Whilst some of the comments offered with respect to the target market may find application in other arenas, it is beyond the scope of this paper to explicitly consider highly heterogeneous markets. Given also the wide scope of this paper key drivers for change and their likely implications for the commercial office property market are identified at a global level (within the above established parameters). Accordingly, the focus is necessarily such that it serves to reflect overarching directions at a universal level (with the effect being that direct applicability to individual markets - when viewed in isolation on a geographic or property type specific basis – may not be fitting in all instances)

Response and damage evaluation of reinforced concrete frames subjected to blast loading

Bomb attacks carried out by terrorists, targeting high occupancy buildings, have become increasingly common in recent times. Large numbers of casualties and property damage result from overpressure of the blast followed by failing of structural elements. Understanding the blast response of multi-storey buildings and evaluating their remaining life have therefore become important. Response and damage analysis of single structural components, such as columns or slabs, to explosive loads have been examined in the literature, but the studies on blast response and damage analysis of structural frames in multi-storey buildings is limited and this is necessary for assessing the vulnerability of them. This paper investigates the blast response and damage evaluation of reinforced concrete (RC) frames, designed for normal gravity loads, in order to evaluate their remaining life. Numerical modelling and analysis were carried out using the explicit finite element software, LS DYNA. The modelling and analysis takes into consideration reinforcement details together and material performance under higher strain rates. Damage indices for columns are calculated based on their residual and original capacities. Numerical results generated in the can be used to identify relationships between the blast load parameters and the column damage. Damage index curve will provide a simple means for assessing the damage to a typical multi-storey building RC frame under an external bomb circumstance.