The sudden switch to distance education to contain the outbreak of the COVID-19 pandemic has fundamentally altered adolescents’ lives around the globe. The present research aims to identify psychological characteristics that relate to adolescents’ well-being in terms of positive emotion and intrinsic learning motivation, and key characteristics of their learning behavior in a situation of unplanned, involuntary distance education. Following Self-Determination Theory, experienced competence, autonomy, and relatedness were assumed to relate to active learning behavior (i.e., engagement and persistence), and negatively relate to passive learning behavior (i.e., procrastination), mediated via positive emotion and intrinsic learning motivation. Data were collected via online questionnaires in altogether eight countries from Europe, Asia, and North America (N = 25,305) and comparable results across countries were expected. Experienced competence was consistently found to relate to positive emotion and intrinsic learning motivation, and, in turn, active learning behavior in terms of engagement and persistence. The study results further highlight the role of perceived relatedness for positive emotion. The high proportions of explained variance speak in favor of taking these central results into account when designing distance education in times of COVID-19.

We investigate the addition of constraints on the function image and its derivatives for the incorporation of prior knowledge in symbolic regression. The approach is called shape-constrained symbolic regression and allows us to enforce e.g. monotonicity of the function over selected inputs. The aim is to find models which conform to expected behaviour and which have improved extrapolation capabilities. We demonstrate the feasibility of the idea and propose and compare two evolutionary algorithms for shapeconstrained symbolic regression: i) an extension of tree-based genetic programming which discards infeasible solutions in the selection step, and ii) a two population evolutionary algorithm that separates the feasible from the infeasible solutions. In both algorithms we use interval arithmetic to approximate bounds for models and their partial derivatives. The algorithms are tested on a set of 19 synthetic and four real-world regression problems. Both algorithms are able to identify models which conform to shape constraints which is not the case for the unmodified symbolic regression algorithms. However, the predictive accuracy of models with constraints is worse on the training set and the test set. Shape-constrained polynomial regression produces the best results for the test set but also significantly larger models.

No abstract

Erik.Sonnhammer@scilifelab.se.