One of the long-standing challenges in Artificial Intelligence for goal-directed behavior is to build a single agent which can solve multiple tasks. Recent progress in multi-task learning for goal-directed sequential tasks has been in the form of distillation based learning wherein a single student network learns from multiple task-specific expert networks by mimicking the task-specific policie...

one of the major criticisms leveled at task-based language teaching (tblt), despite its countless merits, is developing fluency at the cost of accuracy. the post-task stage affords a number of options to counteract this downside through task repetition and task recycling. these two options are considered to positively affect learners' oral performance in terms of fluency, accuracy, and complexi...

An active learning approach is proposed to automatically analyze speech recognition tasks and select particularly useful adaptation data. In this approach, the distribution of task data is first estimated, which is a combination of two distributions based on N-best recognition results and low confidence data. After that, a subset of adaptation data is selected in two stages using a greedy algor...

We extend the classical single-task active learning (AL) approach. In the multi-task active learning (MTAL) paradigm, we select examples for several annotation tasks rather than for a single one as usually done in the context of AL. We introduce two MTAL metaprotocols, alternating selection and rank combination, and propose a method to implement them in practice. We experiment with a twotask an...