Learning-to-Rank with Partitioned Preference: Fast Estimation for the Plackett-Luce Model
14. dubna 2021
Řečníci
Xinyang Yi
Řečník · 0 sledujících
Weijing Tang
Řečník · 0 sledujících
Lichan Hong
Řečník · 0 sledujících
Qiaozhu Mei
Řečník · 0 sledujících
O prezentaci
We consider the problem of listwise learning-to-rank (LTR) on data with \textit{partitioned preference}, where a set of items are sliced into ordered and disjoint partitions, but the ranking of items within a partition is unknown. The Plackett-Luce (PL) model has been widely used in listwise LTR methods. However, given $N$ items with $M$ partitions, calculating the likelihood of data with partitioned preference under the PL model has a time complexity of $O(N+S!)$, where $S$ is the maximum size of the top $M-1$ partitions. This computational challenge restrains existing PL-based listwise LTR methods to only a special case of partitioned preference, \textit{top-$K$ ranking}, where the exact order of the top $K$ items is known. In this paper, we exploit a random utility model formulation of the PL model and propose an efficient approach through numerical integration for calculating the likelihood. This numerical approach reduces the aforementioned time complexity to $O(N+MS)$, which allows training deep-neural-network-based ranking models with a large output space. We demonstrate that the proposed method outperforms well-known LTR baselines and remains scalable through both simulation experiments and applications to real-world eXtreme Multi-Label (XML) classification tasks. The proposed method also achieves state-of-the-art performance on XML datasets with relatively large numbers of labels per sample.We consider the problem of listwise learning-to-rank (LTR) on data with \textit{partitioned preference}, where a set of items are sliced into ordered and disjoint partitions, but the ranking of items within a partition is unknown. The Plackett-Luce (PL) model has been widely used in listwise LTR methods. However, given $N$ items with $M$ partitions, calculating the likelihood of data with partitioned preference under the PL model has a time complexity of $O(N+S!)$, where $S$ is the maximum size…
Organizátor
AISTATS 2021
Účet · 63 sledujících
Kategorie
Počítačová věda a IT
Kategorie · 14,8k prezentací
O organizátorovi (AISTATS 2021)
The 24th International Conference on Artificial Intelligence and Statistics was held virtually from Tuesday, 13 April 2021 to Thursday, 15 April 2021.
Baví vás formát? Nechte SlidesLive zachytit svou akci!
Profesionální natáčení a streamování po celém světě.
Sdílení
Doporučená videa
Prezentace na podobné téma, kategorii nebo přednášejícího
Robust Learning under Strong Noise via SQs
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Differentiating the Value Function by using Convex Duality
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Online Model Selection for Reinforcement Learning with Function Approximation
Zhlédnout později
Jonathan Lee, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Confident Off-Policy Evaluation and Selection through Self-Normalized Importance Weighting
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
On the Generalization Properties of Adversarial Training
Zhlédnout později
Yue Xing, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
On the Absence of Spurious Local Minima in Nonlinear Low-Rank Matrix Recovery Problems
Zhlédnout později
Yingjie Bi, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %