Sequential Optimal Recovery: A Paradigm for Active Learning

In most classical frameworks for learning from examples, it is assumed that examples are randomly drawn and presented to the learner. In this paper, we consider the possibility of a more active learner who is allowed to choose his/her own examples. Our investigations are carried out in a function approximation setting. In particular, using arguments from optimal recovery (Micchelli and Rivlin, 1976), we develop an adaptive sampling strategy (equivalent to adaptive approximation) for arbitrary approximation schemes. We provide a general formulation of the problem and show how it can be regarded as sequential optimal recovery. We demonstrate the application of this general formulation to two special cases of functions on the real line 1) monotonically increasing functions and 2) functions with bounded derivative. An extensive investigation of the sample complexity of approximating these functions is conducted yielding both theoretical and empirical results on test functions. Our theoretical results (stated insPAC-style), along with the simulations demonstrate the superiority of our active scheme over both passive learning as well as classical optimal recovery. The analysis of active function approximation is conducted in a worst-case setting, in contrast with other Bayesian paradigms obtained from optimal design (Mackay, 1992).

Learning from Incomplete Data

Real-world learning tasks often involve high-dimensional data sets with complex patterns of missing features. In this paper we review the problem of learning from incomplete data from two statistical perspectives---the likelihood-based and the Bayesian. The goal is two-fold: to place current neural network approaches to missing data within a statistical framework, and to describe a set of algorithms, derived from the likelihood-based framework, that handle clustering, classification, and function approximation from incomplete data in a principled and efficient manner. These algorithms are based on mixture modeling and make two distinct appeals to the Expectation-Maximization (EM) principle (Dempster, Laird, and Rubin 1977)---both for the estimation of mixture components and for coping with the missing data.

The role of the humanities in the Bologna idea of a university : learning from the american model?. 'El papel de la humanidad en la idea de Bolonia de una universidad : ¿el aprender del modelo americano?'.

Resumen tomado de la publicación. Monográfico con el título: El proceso de Bolonia : dinámicas y desafíos de la enseñanza superior en Europa a comienzos de una nueva época

Primary schools and ICT : learning from pupil perspectives.

El uso de la tecnología de la información y de la comunicación (TIC) en las escuelas primarias es a menudo problemática y frustrante tanto para los profesores como para los alumnos. Este texto se basa en un estudio sobre las experiencias y percepciones de más de seiscientos alumnos de primaria para explorar cómo se puede mejorar el uso de las TIC en ese nivel y hace una serie de sugerencias prácticas para su modificación en las escuelas teniendo en cuenta qué emociona, motiva e interesa a los niños en y fuera del aula.

Toward a methodology for exploring Mathematics preservice teachers' learning from a sociocultural perspective. 'Hacia una metodología para explorar el aprendizaje de futuros profesores de Matemáticas desde una perspectiva sociocultural'.

Resumen basado en el de la publicación

Learning from the past : the legacy of medieval Iberia for 21st century education. 'El legado de la Península Ibérica en la época medieval : una valiosa fuente de recursos para los retos educativos del siglo XXI'.

Resumen basado en el de la publicación

Learning from errors : effects of teachers training on students??? attitudes towards and their individual use of errors

Resumen basado en el de la publicaci??n

The relation of E.Q and I.Q. to learning from the standard self-instructional programs in spelling and english

This paper reviews a study that was done with hearing and hearing impaired children to test the effectiveness of self-instructional programs and whether the results can be correlated with Educational Quotient and Intelligence Quotient.

Learning from risks: A tool for post-project risk assessment

Risk management (RM) comprises of risk identification, risk analysis, response planning, monitoring and action planning tasks that are carried out throughout the life cycle of a project in order to ensure that project objectives are met. Although the methodological aspects of RM are well-defined, the philosophical background is rather vague. In this paper, a learning-based approach is proposed. In order to implement this approach in practice, a tool has been developed to facilitate construction of a lessons learned database that contains risk-related information and risk assessment throughout the life cycle of a project. The tool is tested on a real construction project. The case study findings demonstrate that it can be used for storing as well as updating risk-related information and finally, carrying out a post-project appraisal. The major weaknesses of the tool are identified as, subjectivity of the risk rating process and unwillingness of people to enter information about reasons of failure.