Mind change complexity of learning logic programs

The present paper motivates the study of mind change complexity for learning minimal models of length-bounded logic programs. It establishes ordinal mind change complexity bounds for learnability of these classes both from positive facts and from positive and negative facts. Building on Angluin’s notion of finite thickness and Wright’s work on finite elasticity, Shinohara defined the property of bounded finite thickness to give a sufficient condition for learnability of indexed families of computable languages from positive data. This paper shows that an effective version of Shinohara’s notion of bounded finite thickness gives sufficient conditions for learnability with ordinal mind change bound, both in the context of learnability from positive data and for learnability from complete (both positive and negative) data. Let Omega be a notation for the first limit ordinal. Then, it is shown that if a language defining framework yields a uniformly decidable family of languages and has effective bounded finite thickness, then for each natural number m >0, the class of languages defined by formal systems of length <= m: • is identifiable in the limit from positive data with a mind change bound of Omega (power)m; • is identifiable in the limit from both positive and negative data with an ordinal mind change bound of Omega × m. The above sufficient conditions are employed to give an ordinal mind change bound for learnability of minimal models of various classes of length-bounded Prolog programs, including Shapiro’s linear programs, Arimura and Shinohara’s depth-bounded linearly covering programs, and Krishna Rao’s depth-bounded linearly moded programs. It is also noted that the bound for learning from positive data is tight for the example classes considered.

Mind change optimal learning of Bayes net structure

This paper analyzes the problem of learning the structure of a Bayes net (BN) in the theoretical framework of Gold’s learning paradigm. Bayes nets are one of the most prominent formalisms for knowledge representation and probabilistic and causal reasoning. We follow constraint-based approaches to learning Bayes net structure, where learning is based on observed conditional dependencies between variables of interest (e.g., “X is dependent on Y given any assignment to variable Z”). Applying learning criteria in this model leads to the following results. (1) The mind change complexity of identifying a Bayes net graph over variables V from dependency data is |V| 2 , the maximum number of edges. (2) There is a unique fastest mind-change optimal Bayes net learner; convergence speed is evaluated using Gold’s dominance notion of “uniformly faster convergence”. This learner conjectures a graph if it is the unique Bayes net pattern that satisfies the observed dependencies with a minimum number of edges, and outputs “no guess” otherwise. Therefore we are using standard learning criteria to define a natural and novel Bayes net learning algorithm. We investigate the complexity of computing the output of the fastest mind-change optimal learner, and show that this problem is NP-hard (assuming P = RP). To our knowledge this is the first NP-hardness result concerning the existence of a uniquely optimal Bayes net structure.

On a Syntactic Characterization of Classification with a Mind Change Bound

Most learning paradigms impose a particular syntax on the class of concepts to be learned; the chosen syntax can dramatically affect whether the class is learnable or not. For classification paradigms, where the task is to determine whether the underlying world does or does not have a particular property, how that property is represented has no implication on the power of a classifier that just outputs 1’s or 0’s. But is it possible to give a canonical syntactic representation of the class of concepts that are classifiable according to the particular criteria of a given paradigm? We provide a positive answer to this question for classification in the limit paradigms in a logical setting, with ordinal mind change bounds as a measure of complexity. The syntactic characterization that emerges enables to derive that if a possibly noncomputable classifier can perform the task assigned to it by the paradigm, then a computable classifier can also perform the same task. The syntactic characterization is strongly related to the difference hierarchy over the class of open sets of some topological space; this space is naturally defined from the class of possible worlds and possible data of the learning paradigm.

On Ordinal VC-Dimension and Some Notions of Complexity

We generalize the classical notion of Vapnik–Chernovenkis (VC) dimension to ordinal VC-dimension, in the context of logical learning paradigms. Logical learning paradigms encompass the numerical learning paradigms commonly studied in Inductive Inference. A logical learning paradigm is defined as a set W of structures over some vocabulary, and a set D of first-order formulas that represent data. The sets of models of ϕ in W, where ϕ varies over D, generate a natural topology W over W. We show that if D is closed under boolean operators, then the notion of ordinal VC-dimension offers a perfect characterization for the problem of predicting the truth of the members of D in a member of W, with an ordinal bound on the number of mistakes. This shows that the notion of VC-dimension has a natural interpretation in Inductive Inference, when cast into a logical setting. We also study the relationships between predictive complexity, selective complexity—a variation on predictive complexity—and mind change complexity. The assumptions that D is closed under boolean operators and that W is compact often play a crucial role to establish connections between these concepts. We then consider a computable setting with effective versions of the complexity measures, and show that the equivalence between ordinal VC-dimension and predictive complexity fails. More precisely, we prove that the effective ordinal VC-dimension of a paradigm can be defined when all other effective notions of complexity are undefined. On a better note, when W is compact, all effective notions of complexity are defined, though they are not related as in the noncomputable version of the framework.

Mind change efficient learning

This paper studies efficient learning with respect to mind changes. Our starting point is the idea that a learner that is efficient with respect to mind changes minimizes mind changes not only globally in the entire learning problem, but also locally in subproblems after receiving some evidence. Formalizing this idea leads to the notion of uniform mind change optimality. We characterize the structure of language classes that can be identified with at most α mind changes by some learner (not necessarily effective): A language class L is identifiable with α mind changes iff the accumulation order of L is at most α. Accumulation order is a classic concept from point-set topology. To aid the construction of learning algorithms, we show that the characteristic property of uniformly mind change optimal learners is that they output conjectures (languages) with maximal accumulation order. We illustrate the theory by describing mind change optimal learners for various problems such as identifying linear subspaces and one-variable patterns.

On the data consumption benefits of accepting increased uncertainty

In the context of learning paradigms of identification in the limit, we address the question: why is uncertainty sometimes desirable? We use mind change bounds on the output hypotheses as a measure of uncertainty, and interpret ‘desirable’ as reduction in data memorization, also defined in terms of mind change bounds. The resulting model is closely related to iterative learning with bounded mind change complexity, but the dual use of mind change bounds — for hypotheses and for data — is a key distinctive feature of our approach. We show that situations exists where the more mind changes the learner is willing to accept, the lesser the amount of data it needs to remember in order to converge to the correct hypothesis. We also investigate relationships between our model and learning from good examples, set-driven, monotonic and strong-monotonic learners, as well as class-comprising versus class-preserving learnability.

On the data consumption benefits of accepting increased uncertainty

In the context of learning paradigms of identification in the limit, we address the question: why is uncertainty sometimes desirable? We use mind change bounds on the output hypotheses as a measure of uncertainty and interpret ‘desirable’ as reduction in data memorization, also defined in terms of mind change bounds. The resulting model is closely related to iterative learning with bounded mind change complexity, but the dual use of mind change bounds — for hypotheses and for data — is a key distinctive feature of our approach. We show that situations exist where the more mind changes the learner is willing to accept, the less the amount of data it needs to remember in order to converge to the correct hypothesis. We also investigate relationships between our model and learning from good examples, set-driven, monotonic and strong-monotonic learners, as well as class-comprising versus class-preserving learnability.

Compute inclusion depth of a pattern

We define a concept of inclusion depth (see Definition 1) to capture mind-change complexity [3,1] of pattern identification problems [2]. Our basic question is whether the inclusion depth for any pattern is computable. We conjecture a combinatorial characterization that, if true, leads to a linear time algorithm to compute inclusion depth.

Individuals' Responsibility in Corporations

I discuss role responsibly, individual responsibility and collective responsibility in corporate multinational setting. My case study is about minerals used in electronics that come from the Democratic Republic of Congo. What I try to show throughout the thesis is how many things need to be taken into consideration when we discuss the responsibility of individuals in corporations. No easy and simple answers are available. Instead, we must keep in mind the complexity of the situation at all times, judging cases on individual basis, emphasizing the importance of individual judgement and virtue, as well as the responsibility we all share as members of groups and the wider society. I begin by discussing the demands that are placed on us as employees. There is always a potential for a conflict between our different roles and also the wider demands placed on us. Role demands are usually much more specific than the wider question of how we should act as human beings. The terminology of roles can also be misleading as it can create illusions about our work selves being somehow radically separated from our everyday, true selves. The nature of collective decision-making and its implications for responsibility is important too. When discussing the moral responsibility of an employee in a corporate setting, one must take into account arguments from individual and collective responsibility, as well as role ethics. Individual responsibility is not a separate or competing notion from that of collective responsibility. Rather, the two are interlinked. Individuals' responsibilities in collective settings combine both individual responsibility and collective responsibility (which is different from aggregate individual responsibility). In the majority of cases, both will apply in various degrees. Some members might have individual responsibility in addition to the collective responsibility, while others just the collective responsibility. There are also times when no-one bears individual moral responsibility but the members are still responsible for the collective part. My intuition is that collective moral responsibility is strongly linked to the way the collective setting affects individual judgements and moulds the decisions, and how the individuals use the collective setting to further their own ends. Individuals remain the moral agents but responsibility is collective if the actions in question are collective in character. I also explore the impacts of bureaucratic ethic and its influence on the individual. Bureaucracies can compartmentalize work to such a degree that individual human action is reduced to mere behaviour. Responsibility is diffused and the people working in the bureaucracy can come to view their actions to be outside the normal human realm where they would be responsible for what they do. Language games and rules, anonymity, internal power struggles, and the fragmentation of information are just some of the reasons responsibility and morality can get blurry in big institutional settings. Throughout the thesis I defend the following theses: ● People act differently depending on their roles. This is necessary for our society to function, but the more specific role demands should always be kept in check by the wider requirements of being a good human being. ● Acts in corporations (and other large collectives) are not reducible to individual actions, and cannot be explained fully by the behaviour of individual employees. ● Individuals are responsible for the actions that they undertake in the collective as role occupiers and are very rarely off the hook. Hiding behind role demands is usually only an excuse and shows a lack of virtue. ● Individuals in roles can be responsible even when the collective is not. This depends on if the act they performed was corporate in nature or not. ● Bureaucratic structure affects individual thinking and is not always a healthy environment to work in. ● Individual members can share responsibility with the collective and our share of the collective responsibility is strongly linked to our relations. ● Corporations and other collectives can be responsible for harm even when no individual is at fault. The structure and the policies of the collective are crucial. ● Socialization plays an important role in our morality at both work and outside it. We are all responsible for the kind of moral context we create. ● When accepting a role or a position in a collective, we are attaching ourselves with the values of that collective. ● Ethical theories should put more emphasis on good judgement and decision-making instead of vague generalisations. My conclusion is that the individual person is always in the centre when it comes to responsibility, and not so easily off the hook as we sometimes think. What we do, and especially who we choose to associate ourselves with, does matter and we should be more careful when we choose who we work for. Individuals within corporations are responsible for choosing that the corporation they associate with is one that they can ascribe to morally, if not fully, then at least for the most part. Individuals are also inclusively responsible to a varying degree for the collective activities they contribute to, even in overdetermined contexts. We all are responsible for the kind of corporations we choose to support through our actions as consumers, investors and citizens.

Lenguaje y Teoría de la mente: una aproximación multidimensional

El propósito de este trabajo es realizar una puesta al día sobre las relaciones entre el desarrollo del lenguaje y el desarrollo de la Teoría de la Mente. Entre las hipótesis propuestas para conceptualizar dichas relaciones, realizamos un examen más exhaustivo de aquellos modelos que apoyan una implicación directa entre lenguaje y Teoría de la Mente. En este contexto, en primer lugar subrayamos la necesidad de ampliar dicha noción para incluir habilidades anteriores y posteriores a la comprensión de la falsa creencia. En segundo lugar, examinamos las diferentes hipótesis acerca de los aspectos del lenguaje más vinculados al desarrollo sociocognitivo, y las principales evidencias empíricas que las sustentan. Finalmente, discutimos un modelo de causalidad recíproca donde los vínculos entre lenguaje y Teoría de la Mente variarán según el momento del desarrollo considerado

As atitudes dos professores face à inclusão de alunos com deficiência no Ensino Regular

As sociedades actuais caracterizam-se pela mudança, complexidade e diversidade cultural. De dia para dia assistimos cada vez mais à passagem das responsabilidades educativas da comunidade para a escola, assumindo esta, um papel preponderante na formação dos nossos jovens. O professor, é assim, considerado um elemento chave na construção de uma escola inclusiva e para que haja uma efectiva mudança da escola e das estratégias da sala de aula (Costa, 1999). O objectivo deste estudo é investigar as atitudes dos professores, determinando como estes percepcionam as suas Competências e de que forma essas atitudes são influenciadas, ou não, tendo em conta a experiência profissional, o grupo disciplinar a que pertencem, a experiência no ensino de alunos com deficiência e a formação inicial cujas áreas de estudo, estiveram directamente relacionados com a deficiência. A amostra deste estudo foi constituída por um total de 741 professores do ensino regular (534 afectos ao género feminino e 207 afectos ao género masculino). Foi aplicado o questionário APIAD – Atitudes dos Professores face à Inclusão de Alunos com Deficiência (Leitão, 2011), inferido através da resposta a 14 afirmações, em que cada uma delas deverá ser relacionado com 4 condições de deficiência (deficiência visual, deficiência auditiva, deficiência mental e deficiência motora). Concluiu-se então que, relativamente ao Grupo Disciplinar, as diferenças significativas se encontram entre os professores de Educação Física e os professores de Humanidades e de Ciências, sendo que os primeiros têm uma atitude mais positiva relativamente às suas competências. Quanto ao Contacto com a Deficiência, os resultados indicaram que os professores que têm experiência no ensino de alunos com deficiência têm uma atitude mais positiva relativamente às suas competências. No que se refere à Experiência Profissional, os nossos resultados encontram diferenças significativas entre os professores com menos anos de experiência e os professores com mais anos de experiência, sendo que os primeiros têm uma atitude mais positiva relativamente às suas competências Por último, em relação à Formação Inicial, os professores cujas áreas de estudo, na sua formação inicial, estiveram directamente relacionados com deficiência têm uma atitude mais positiva relativamente às suas competências. ABSTRACT: Today, societies are characterized by change, complexity and cultural diversity. Each day, we saw more and more responsibilities passing from the educational community to school, assuming that a major role in the formation of our youth. Teacher is considered a key element in the building of an inclusive school and for that there is a real change of school and classroom strategies (Costa, 1999). The purpose of this study is to investigate teacher´s attitudes, determining how they perceive their skills and how these attitudes are or aren´t influenced taking into account the professional experience, the subject group they belong to, the experience in students with those particular needs teaching and last, the initial training with chairs that had contact with people with needs. The study sample consisted in a total of 741 teachers of the regular education (534 allocated to the female and 207 assigned to the male gender). We used the questionnaire APIAD - Attitudes of teachers towards the Inclusion of Students with Disabilities (Leitão, 2011), inferred by the response to 14 statements, each one of them should be related to four conditions of disability (visual impairment, hearing impairment, mental retardation and physical disabilities). It was concluded then that for the Disciplinary Group, significant differences are found between the physical education teachers and teachers of Humanities and Sciences, and the former have a more positive attitude to their skills. How to contact the Disability, the results indicated that teachers who have experience teaching students with disabilities have a more positive attitude to their skills. With regard to professional experience, significant differences were found between teachers with fewer years of experience and teachers with more years of experience, and the former have a more positive attitude to their skills. Finally, regarding to initial training, teachers in their initial training formation who have contact with people with disabilities have a more positive attitude to their skills.

Sharpening the focus of force field analysis

The purpose is to explore the inherent complexity of Kurt Lewin's force field theory through applied analysis of organizational case examples and related methods. The methodology applies a range of tools from the consultancy research domain, including force field analysis of complex organizational scenarios, and applies bricolage and corroboration to emerging discoveries from semi-structured interviews, author experience, critical reflection and literature survey. Findings are that linear representation of internal and external forces in organizational applications of field theory does not fully explain the paradox of inverse vectors in the forces of change. The force field is not an impermeable thing; instead, it morphs. Examples of the inverse principle and its effects are detailed and extended in this analysis. The implications of the research are that force field analysis and related change processes promoted in organizational change literature run the risk of missing key complexities. The inclusion of the inverse principle can provide enhanced, holistic understanding of the prevailing forces for change. The augmentation of the early work of Kurt Lewin, and extension of previous analyses of his legacy in the Journal of Change Management and elsewhere, provide, in this article, change analysis insights that align well with current organizational environments.

A influência do estilo de liderança na construção de relações de confiança em uma instituição financeira

A turbulência dos mercados globais, o acirramento da concorrência, a velocidade das mudanças, a complexidade e a incerteza do cenário político, econômico e social trazem incessantes desafios para as organizações, que buscam responder a esses movimentos dinâmicos com a celeridade necessária, a fim de obter vantagem competitiva e assegurar a sustentabilidade dos negócios. Nessa perspectiva, a liderança tem um papel fundamental por ser capaz de mobilizar e orientar a força de trabalho para o cumprimento das metas organizacionais, uma vez que o líder pode influenciar positivamente e inspirar pessoas a dar o melhor de si em benefício da organização. O engajamento, a motivação e a cooperação que propiciam essa entrega pessoal ocorrem a partir da percepção dos liderados a respeito do comportamento do líder exemplar, o que remete à construção de vínculos de confiança. Em função disso, este trabalho teve como objetivo identificar o estilo de liderança prevalecente em uma unidade de uma instituição financeira, bem como avaliar se este estilo propicia a construção de relações de confiança entre líder e liderados. Para tanto, inicialmente foi realizado um levantamento bibliográfico sobre os temas liderança e confiança. Na sequência, foram aplicados os questionários Multifactor Leadership Questionnaire (MLQ) e Behavioral Trust Inventory (BTI) e realizadas entrevistas, a fim de obter os dados para análise. Como resultado da pesquisa concluiu-se que prevalece na unidade estudada um estilo com traços da liderança transformacional e da liderança transacional recompensa contingente. Ficou evidenciado que esse estilo de liderança constrói vínculos de confiança mútua, demonstrando que as ações do líder fomentam o reconhecimento de sua credibilidade e auxiliam na solução de dilemas de gestão, possibilitando a cooperação espontânea para o alcance dos objetivos estratégicos da instituição.