No abstract

No abstract

We present a simple and scalable algorithm for top-N recommen-\ud dation able to deal with very large datasets and (binary rated) im-\ud plicit feedback. We focus on memory-based collaborative filtering\ud algorithms similar to the well known neighboor based technique\ud for explicit feedback. The major difference, that makes the algo-\ud rithm particularly scalable, is that it uses positive feedback only\ud and no explicit computation of the complete (user-by-user or item-\ud by-item) similarity matrix needs to be performed.\ud The study of the proposed algorithm has been conducted on data\ud from the Million Songs Dataset (MSD) challenge whose task was\ud to suggest a set of songs (out of more than 380k available songs)\ud to more than 100k users given half of the user listening history and\ud complete listening history of other 1 million people.\ud In particular, we investigate on the entire recommendation pipeline,\ud starting from the definition of suitable similarity and scoring func-\ud tions and suggestions on how to aggregate multiple ranking strate-\ud gies to define the overall recommendation. The technique we are\ud proposing extends and improves the one that already won the MSD\ud challenge last year

Abstract. Recent results in theoretical machine learning seem to suggest that nice properties of the margin distribution over a training set turns out in a good performance of a classifier. The same principle has been already used in SVM and other kernel based methods as the associated optimization problems try to maximize the minimum of these margins. In this paper, we propose a kernel based method for the direct optimization of the margin distribution (KM-OMD). The method is motivated and analyzed from a game theoretical perspective. A quite efficient optimization algorithm is then proposed. Experimental results over a standard benchmark of 13 datasets have clearly shown state-of-the-art performances.

Recent literature has shown the merits of having deep representations in the context of neural networks. An emerging challenge in kernel learning is the definition of similar deep representations. In this paper, we propose a general methodology to define a hierarchy of base kernels with increasing expressiveness and combine them via multiple kernel learning (MKL) with the aim to generate overall deeper kernels. As a leading example, this methodology is applied to learning the kernel in the space of Dot-Product Polynomials (DPPs), that is a positive combination of homogeneous polynomial kernels (HPKs). We show theoretical properties about the expressiveness of HPKs that make their combination empirically very effective. This can also be seen as learning the coefficients of the Maclaurin expansion of any definite positive dot product kernel thus making our proposed method generally applicable. We empirically show the merits of our approach comparing the effectiveness of the kernel generated by our method against baseline kernels (including homogeneous and non homogeneous polynomials, RBF, etc...) and against another hierarchical approach on several benchmark datasets.