In the university timetabling problem, sometimes additions or cancellations of course sections occur shortly before the beginning of the academic term, necessitating last-minute teaching staffing changes. We present a decision-making framework that both minimizes the number of course swaps, which are inconvenient to faculty members, and maximizes faculty members' preferences for times they wish to teach. The model is formulated as an integer linear program (ILP). Numerical simulations for a hypothetical mid-sized academic department are presented.

In recent years, deep learning approaches have gained significant interest as a way of building hierarchical representations from unlabeled data. However, to our knowledge, these deep learning approaches have not been extensively studied for auditory data. In this paper, we apply convolutional deep belief networks to audio data and empirically evaluate them on various audio classification tasks. For the case of speech data, we show that the learned features correspond to phones/phonemes. In addition, our feature representations trained from unlabeled audio data show very good performance for multiple audio classification tasks. We hope that this paper will inspire more research on deep learning approaches applied to a wide range of audio recognition tasks.