scholarly journals PairRank: Online Pairwise Learning to Rank by Divide-and-Conquer

Author(s):  
Yiling Jia ◽  
Huazheng Wang ◽  
Stephen Guo ◽  
Hongning Wang
2019 ◽  
Vol 18 (01) ◽  
pp. 109-127
Author(s):  
Ting Hu ◽  
Jun Fan ◽  
Dao-Hong Xiang

In this paper, we establish the error analysis for distributed pairwise learning with multi-penalty regularization, based on a divide-and-conquer strategy. We demonstrate with [Formula: see text]-error bound that the learning performance of this distributed learning scheme is as good as that of a single machine which could process the whole data. With semi-supervised data, we can relax the restriction of the number of local machines and enlarge the range of the target function to guarantee the optimal learning rate. As a concrete example, we show that the work in this paper can apply to the distributed pairwise learning algorithm with manifold regularization.


IEEE Access ◽  
2020 ◽  
Vol 8 ◽  
pp. 192352-192367
Author(s):  
Yiqing Shi ◽  
Yuzhen Niu ◽  
Wenzhong Guo ◽  
Yize Huang ◽  
Jiamei Zhan

2020 ◽  
Vol 34 (05) ◽  
pp. 9073-9080
Author(s):  
Ming Tu ◽  
Kevin Huang ◽  
Guangtao Wang ◽  
Jing Huang ◽  
Xiaodong He ◽  
...  

Interpretable multi-hop reading comprehension (RC) over multiple documents is a challenging problem because it demands reasoning over multiple information sources and explaining the answer prediction by providing supporting evidences. In this paper, we propose an effective and interpretable Select, Answer and Explain (SAE) system to solve the multi-document RC problem. Our system first filters out answer-unrelated documents and thus reduce the amount of distraction information. This is achieved by a document classifier trained with a novel pairwise learning-to-rank loss. The selected answer-related documents are then input to a model to jointly predict the answer and supporting sentences. The model is optimized with a multi-task learning objective on both token level for answer prediction and sentence level for supporting sentences prediction, together with an attention-based interaction between these two tasks. Evaluated on HotpotQA, a challenging multi-hop RC data set, the proposed SAE system achieves top competitive performance in distractor setting compared to other existing systems on the leaderboard.


Author(s):  
Yuan Liu ◽  
Tao Mei ◽  
Jinhui Tang ◽  
Xiuqing Wu ◽  
Xian-Sheng Hua

2021 ◽  
pp. 1-12
Author(s):  
Wang Zhou ◽  
Yujun Yang ◽  
Yajun Du ◽  
Amin Ul Haq

Recent researches indicate that pairwise learning to rank methods could achieve high performance in dealing with data sparsity and long tail distribution in item recommendation, although suffering from problems such as high computational complexity and insufficient samples, which may cause low convergence and inaccuracy. To further improve the performance in computational capability and recommendation accuracy, in this article, a novel deep neural network based recommender architecture referred to as PDLR is proposed, in which the item corpus will be partitioned into two collections of positive instances and negative items respectively, and pairwise comparison will be performed between the positive instances and negative samples to learn the preference degree for each user. With the powerful capability of neural network, PDLR could capture rich interactions between each user and items as well as the intricate relations between items. As a result, PDLR could minimize the ranking loss, and achieve significant improvement in ranking accuracy. In practice, experimental results over four real world datasets also demonstrate the superiority of PDLR in contrast to state-of-the-art recommender approaches, in terms of Rec@N, Prec@N, AUC and NDCG@N.


2013 ◽  
Vol 29 (22) ◽  
pp. 2909-2917 ◽  
Author(s):  
R. Leaman ◽  
R. Islamaj Dogan ◽  
Z. Lu

Sign in / Sign up

Export Citation Format

Share Document