Maximum-Margin Matrix Factorization
نویسندگان
چکیده
We present a novel approach to collaborative prediction, using low-norm instead of low-rank factorizations. The approach is inspired by, and has strong connections to, large-margin linear discrimination. We show how to learn low-norm factorizations by solving a semi-definite program, and discuss generalization error bounds for them.
منابع مشابه
Sparse Learning via Maximum Margin Matrix Factorization
In this paper, an algorithm for sparse learning via Maximum Margin Matrix Factorization(MMMF) is proposed. The algorithm is based on L1 penality and Alternating Direction Method of Multipliers. It shows that with sparse factors, sparse factors method can obtain result as good as dense factors.
متن کاملMaximum Margin Matrix Factorization with Netflix Data
Maximum Margin Matrix Factorization (MMMF), a collaborative filtering method, was recently introduced in [7] followed by an iterative solution presented in [6]. In this paper we analyze the performance of MMMF on a subset of the Netflix data based on RMSE and classification rate. We also present several modifications to improve the performance of the algorithm on the Netflix problem.
متن کاملMax-Margin Semi-NMF
In this paper, we propose a maximum-margin framework for classification using Nonnegative Matrix Factorization. In contrast to previous approaches where the classification and matrix factorization stages are separated, we incorporate the maximum margin constraints within the NMF formulation, i.e we solve for a base matrix that maximizes the margin of the classifier in the low dimensional featur...
متن کاملCOFI RANK - Maximum Margin Matrix Factorization for Collaborative Ranking
In this paper, we consider collaborative filtering as a ranking problem. We present a method which uses Maximum Margin Matrix Factorization and optimizes ranking instead of rating. We employ structured output prediction to optimize for specific non-uniform ranking scores. Experimental results show that our method gives very good ranking scores and scales well on collaborative filtering tasks.
متن کاملMax-margin Non-negative Matrix Factorization
In this paper we introduce a supervised, maximum margin framework for linear and non-linear Non-negative Matrix Factorization. By contrast to existing methods in which the matrix factorization phase (i.e. the feature extraction phase) and the classification phase are separated, we incorporate the maximum margin classification constraints within the NMF formulation. This results to a non-convex ...
متن کاملData Augmented Maximum Margin Matrix Factorization for Flickr Group Recommendation
User groups on photo sharing websites, such as Flickr, are self-organized communities to share photos and conversations with similar interest and have gained massive popularity. However, the huge volume of groups brings troubles for users to decide which group to choose. Further, directly applying collaborative filtering techniques to group recommendation will suffer from cold start problem sin...
متن کامل