Multitask learning without label correspondences

We propose an algorithm to perform multitask learning where each task has potentially distinct label sets and label correspondences are not readily available. This is in contrast with existing methods which either assume that the label sets shared by different tasks are the same or that there exists a label mapping oracle. Our method directly maximizes the mutual information among the labels, and we show that the resulting objective function can be efficiently optimized using existing algorithms. Our proposed approach has a direct application for data integration with different label spaces, such as integrating Yahoo! and DMOZ web directories.

Inclusive design: Reflections on design practice

To investigate the practice of inclusive design in an industrial context and to gain an insight into the industrial perspectives, eight UK design consultancies' participation of the DBA design Challenges were reviewed through formal interviews. It is found that progress has been made in raising inclusive design awareness. However, some useful practices such as the user involvement in the design process is found not feasible in real situation, largely because of the often tight schedule and the complexity of the task. Consequently effective ways of capturing user information needs exploration and accessible design support tools need to be provided, through working with designers.