Recasting transfer as a socio-personal process of adaptable learning

Enabling learning for all through adaptable personal learning environments

Recasting Gradient-Based Meta-Learning as Hierarchical Bayes

Meta-learning allows an intelligent agent to leverage prior learning episodes as a basis for quickly improving performance on a novel task. Bayesian hierarchical modeling provides a theoretical framework for formalizing meta-learning as inference for a set of parameters that are shared across tasks. Here, we reformulate the model-agnostic meta-learning algorithm (MAML) of Finn et al. (2017) as ...

Modelling Learning as a Process

In this paper we present a formal basis for the development of intelligent learning environments that are capable of reasoning about the process of learning and of changing the environment in ways that create possibilities for processes of interaction that increase learning over time. To support this kind of reasoning we have developed a formal account of interaction and of time-extended proces...

A proposal for an adaptable personal learning environment to support learners needs and preferences

In recent years, Virtual Learning Environments (VLE) (also known as Learning Management Systems) have become pervasive in Higher Education. In addition, the explosion in social software and Web 2.0 technologies raises learners’ expectations of flexibility and personalisation, and highlights the rigidity and limitations of the VLE. Recognition of these limitations has given rise to the evolution...

Adaptation in Adaptable Personal Information Environment

In order to support knowledge workers during their tasks of searching, locating and manipulating information, a system that provides information suitable for a particular user’s needs, and that is also able to facilitate annotation, sharing and reuse information is essential. This paper presents Adaptable Personal Information Environment (a-PIE); a service-oriented framework using Open Hypermed...