• Publications
  • Influence
LightGBM: A Highly Efficient Gradient Boosting Decision Tree
TLDR
It is proved that, since the data instances with larger gradients play a more important role in the computation of information gain, GOSS can obtain quite accurate estimation of the information gain with a much smaller data size, and is called LightGBM.
Learning to rank: from pairwise approach to listwise approach
TLDR
It is proposed that learning to rank should adopt the listwise approach in which lists of objects are used as 'instances' in learning, and introduces two probability models, respectively referred to as permutation probability and top k probability, to define a listwise loss function for learning.
Learning to rank for information retrieval
TLDR
Three major approaches to learning to rank are introduced, i.e., the pointwise, pairwise, and listwise approaches, the relationship between the loss functions used in these approaches and the widely-used IR evaluation measures are analyzed, and the performance of these approaches on the LETOR benchmark datasets is evaluated.
MASS: Masked Sequence to Sequence Pre-training for Language Generation
TLDR
This work proposes MAsked Sequence to Sequence pre-training (MASS) for the encoder-decoder based language generation tasks, which achieves the state-of-the-art accuracy on the unsupervised English-French translation, even beating the early attention-based supervised model.
FastSpeech: Fast, Robust and Controllable Text to Speech
TLDR
A novel feed-forward network based on Transformer to generate mel-spectrogram in parallel for TTS is proposed, which speeds up mel-Spectrogram generation by 270x and the end-to-end speech synthesis by 38x and is called FastSpeech.
LETOR: Benchmark Dataset for Research on Learning to Rank for Information Retrieval
TLDR
This paper has constructed a benchmark dataset referred to as LETOR, derived the LETOR data from the existing data sets widely used in IR, namely, OHSUMED and TREC data and provided the results of several state-ofthe-arts learning to rank algorithms on the data.
Listwise approach to learning to rank: theory and algorithm
TLDR
A sufficient condition on consistency for ranking is given, which seems to be the first such result obtained in related research, and analysis on three loss functions: likelihood loss, cosine loss, and cross entropy loss are conducted.
LETOR: A benchmark collection for research on learning to rank for information retrieval
TLDR
The details of the LETOR collection are described and it is shown how it can be used in different kinds of researches, and several state-of-the-art learning to rank algorithms on LETOR are compared.
Learning to Rank for Information Retrieval
The usual approach to optimisation, of ranking algorithms for search and in many other contexts, is to obtain some training set of labeled data and optimise the algorithm on this training set, then
Dual Learning for Machine Translation
TLDR
Experiments show that dual-NMT works very well on English ↔ French translation; especially, by learning from monolingual data, it achieves a comparable accuracy to NMT trained from the full bilingual data for the French-to-English translation task.
...
...