Pairwise Learning to Rank by Neural Networks Revisited: Reconstruction, Theoretical Analysis and Practical Performance

Köppel, Marius; Segner, Alexander; Wagener, Martin; Pensel, Lukas; Karwath, Andreas; Krämer, Stefan

doi:10.48550/arxiv.1909.02768

Cited by 2 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While the authors in (Damke & Hüllermeier, 2021) propose the family of so-called RankGNNs, their competitors are graphs. (Rigutini et al, 2011) applies a neural network approach for preference learning, (Köppel et al, 2019) generalizes (Burges et al, 2005), but these methods require queries as input, which solve a different problem from ours. (Maurya et al, 2021) proposes the first GNN-based model to approximate betweenness and closeness centrality, facilitating locating influential nodes in the graphs in terms of information spread and connectivity.…”

Section: Related Workmentioning

confidence: 99%

GNNRank: Learning Global Rankings from Pairwise Comparisons via Directed Graph Neural Networks

He¹,

Gan²,

Wipf³

et al. 2022

Preprint

View full text Add to dashboard Cite

Recovering global rankings from pairwise comparisons is an important problem with many applications, ranging from time synchronization to sports team ranking. Pairwise comparisons corresponding to matches in a competition can naturally be construed as edges in a directed graph (digraph), whose nodes represent competitors with an unknown rank or skill strength. However, existing methods addressing the rank estimation problem have thus far not utilized powerful neural network architectures to optimize ranking objectives. Hence, we aim to augment a certain ranking algorithm with neural networks, in particular graph neural networks (GNN) for its intrinsic connection to the problem at hand. In this paper, we introduce GNNRank, a modeling framework that is compatible with any GNN capable of learning digraph embeddings, and we devise trainable objectives to encode ranking upsets/violations. This framework includes a ranking score estimation approach, and adds a useful inductive bias by unfolding the Fiedler vector computation of the graph constructed from a learnable similarity matrix. Experimental results on a wide range of data sets show that our methods attain competitive and often superior performance compared with existing approaches. It also shows promising transfer ability to new data based on the trained GNN.

show abstract

Section: Related Workmentioning

confidence: 99%

GNNRank: Learning Global Rankings from Pairwise Comparisons via Directed Graph Neural Networks

He¹,

Gan²,

Wipf³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Learning to Rank formulations for answer selection in QA systems is common practice, most frequently relying on pointwise ranking models (Severyn and Moschitti, 2015;Garg et al, 2019). Our use of discriminative re-ranking (Collins and Koo, 2005) with softmax loss is closer to learning a pairwise ranking by maximizing the multiclass margin between correct and incorrect answers (Joachims, 2002;Burges et al, 2005;Köppel et al, 2019). This is an important distinction from TREC-style answer selection as our ST-generated candidate responses have lower semantic, syntactic, and lexical variance, making pointwise methods less effective.…”

Section: Related Workmentioning

confidence: 99%

Fluent Response Generation for Conversational Question Answering

Baheti¹,

Ritter²,

Small³

2020

Preprint

View full text Add to dashboard Cite

Question answering (QA) is an important aspect of open-domain conversational agents, garnering specific research focus in the conversational QA (ConvQA) subtask. One notable limitation of recent ConvQA efforts is the response being answer span extraction from the target corpus, thus ignoring the natural language generation (NLG) aspect of high-quality conversational agents. In this work, we propose a method for situating QA responses within a SEQ2SEQ NLG approach to generate fluent grammatical answer responses while maintaining correctness. From a technical perspective, we use data augmentation to generate training data for an end-to-end system. Specifically, we develop Syntactic Transformations (STs) to produce question-specific candidate answer responses and rank them using a BERT-based classifier . Human evaluation on SQuAD 2.0 data (Rajpurkar et al., 2018) demonstrate that the proposed model outperforms baseline CoQA and QuAC models in generating conversational responses. We further show our model's scalability by conducting tests on the CoQA dataset. 1

show abstract

Pairwise Learning to Rank by Neural Networks Revisited: Reconstruction, Theoretical Analysis and Practical Performance

Cited by 2 publications

References 0 publications

GNNRank: Learning Global Rankings from Pairwise Comparisons via Directed Graph Neural Networks

GNNRank: Learning Global Rankings from Pairwise Comparisons via Directed Graph Neural Networks

Fluent Response Generation for Conversational Question Answering

Contact Info

Product

Resources

About