Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Fast Greedy MAP Inference for Determinantal Point Process to Improve Recommendation Diversity

About

The determinantal point process (DPP) is an elegant probabilistic model of repulsion with applications in various machine learning tasks including summarization and search. However, the maximum a posteriori (MAP) inference for DPP which plays an important role in many applications is NP-hard, and even the popular greedy algorithm can still be too computationally expensive to be used in large-scale real-time scenarios. To overcome the computational challenge, in this paper, we propose a novel algorithm to greatly accelerate the greedy MAP inference for DPP. In addition, our algorithm also adapts to scenarios where the repulsion is only required among nearby few items in the result sequence. We apply the proposed algorithm to generate relevant and diverse recommendations. Experimental results show that our proposed algorithm is significantly faster than state-of-the-art competitors, and provides a better relevance-diversity trade-off on several public datasets, which is also confirmed in an online A/B test.

Laming Chen, Guoxin Zhang, Hanning Zhou• 2017

Related benchmarks

TaskDatasetResultRank
End-to-end generationASQA
Recall47.07
26
End-to-end generationQAMPARI
Precision19.42
26
Multi-objective Re-rankingML 1M
HR@556.57
13
Multi-objective Re-rankingGrocery
HR@527.12
13
Multi-objective Re-rankingBeauty
Hit Rate @ 522.73
13
RecommendationAMAZON
R@5012.83
11
RecommendationTwitter
R@504.67
11
RecommendationWeibo
R@509.63
11
Route RecommendationMSDR
HR@138.34
8
Route RecommendationProprietary Route Recommendation Dataset (offline)
HR@160.55
8
Showing 10 of 10 rows

Other info

Follow for update