IPAB workshop – 02/03/2017
“Efficient Preference Elicitation with Coarse Preferences”
“We present a new model for the representation of preferences that are based on coarse criteria, leading to better recommendations in problems where user behaviour is consistent with them making decisions by aggregating many individual options. When this is the case, our ‘coarse preference’ model enables an elicitation procedure that outperforms the state-of-the-art utility elicitation procedure that is based on the assumption of additive independence of variables. We demonstrate this through experiments performed with a real-world data set derived from a mobile clothes shopping app. We show that our proposed model achieves double the recommendation acceptance rate and a significant increase in platform profit, due to the use of a decomposition that is better suited to the true nature of the users’ indifference regions. Furthermore, our model is of lower complexity than similar additive independence models, with computation time rising at a slower rate with respect to dimensionality of item descriptions – achieving a halving of computation time in our experiment. Our results have insights for recommendation systems designers, potentially enabling new ways of representing user utility when secondary criteria need to be maximised without compromising performance.”
”End-to-end training of object class detectors for mean average precision”
I will present a method for training CNN-based object class detectors directly using mean average precision (mAP) as the training loss, in a truly end-to-end fashion that includes non-maximum suppresion (NMS) at training time. mAP following NMS forms a piecewise-constant structured loss over thousands of windows, with gradients that do not convey useful information for gradient descent. Hence, we define new, general gradient-like quantities for piecewise constant functions, which have wide applicability. We describe how to calculate these efficiently for mAP following NMS, enabling us to train a detector based on Fast R-CNN directly for mAP.