TY - GEN
T1 - Adversarial training for community question answer selection based on multi-scale matching
AU - Yang, Xiao
AU - Khabsa, Madian
AU - Wang, Miaosen
AU - Wang, Wei
AU - Awadallah, Ahmed Hassan
AU - Kifer, Daniel
AU - Lee Giles, C.
N1 - Funding Information:
We framed the community question answer selection task as a binary classification problem, and presented an adversarial training strategy to alleviate the label imbalance problem. A generative model is introduced to produce challenging negative samples in order to improve the performance of a discriminator. Furthermore, we proposed a Multi-scale Matching model which is enforced to examine context information at different levels of granularities.The proposed method is evaluated on SemEval 2016 and 2017 datasets and achieved state-of-the-art or similar performance. Future work would investigate the stability of GAN training which remains an open research question, especially when discrete sampling is involved. Acknowledgements: We gratefully acknowledge partial support from NSF grant CCF 1317560 and a hardware grant from NVIDIA. This work initiated during Xiao Yang’s internship at Microsoft.
Publisher Copyright:
© 2019, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
PY - 2019
Y1 - 2019
N2 - Community-based question answering (CQA) websites represent an important source of information. As a result, the problem of matching the most valuable answers to their corresponding questions has become an increasingly popular research topic. We frame this task as a binary (relevant/irrelevant) classification problem, and present an adversarial training framework to alleviate label imbalance issue. We employ a generative model to iteratively sample a subset of challenging negative samples to fool our classification model. Both models are alternatively optimized using REINFORCE algorithm. The proposed method is completely different from previous ones, where negative samples in training set are directly used or uniformly down-sampled. Further, we propose using Multi-scale Matching which explicitly inspects the correlation between words and ngrams of different levels of granularity. We evaluate the proposed method on SemEval 2016 and SemEval 2017 datasets and achieves state-of-the-art or similar performance.
AB - Community-based question answering (CQA) websites represent an important source of information. As a result, the problem of matching the most valuable answers to their corresponding questions has become an increasingly popular research topic. We frame this task as a binary (relevant/irrelevant) classification problem, and present an adversarial training framework to alleviate label imbalance issue. We employ a generative model to iteratively sample a subset of challenging negative samples to fool our classification model. Both models are alternatively optimized using REINFORCE algorithm. The proposed method is completely different from previous ones, where negative samples in training set are directly used or uniformly down-sampled. Further, we propose using Multi-scale Matching which explicitly inspects the correlation between words and ngrams of different levels of granularity. We evaluate the proposed method on SemEval 2016 and SemEval 2017 datasets and achieves state-of-the-art or similar performance.
UR - http://www.scopus.com/inward/record.url?scp=85083742723&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85083742723&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85083742723
T3 - 33rd AAAI Conference on Artificial Intelligence, AAAI 2019, 31st Innovative Applications of Artificial Intelligence Conference, IAAI 2019 and the 9th AAAI Symposium on Educational Advances in Artificial Intelligence, EAAI 2019
SP - 395
EP - 402
BT - 33rd AAAI Conference on Artificial Intelligence, AAAI 2019, 31st Innovative Applications of Artificial Intelligence Conference, IAAI 2019 and the 9th AAAI Symposium on Educational Advances in Artificial Intelligence, EAAI 2019
PB - AAAI press
T2 - 33rd AAAI Conference on Artificial Intelligence, AAAI 2019, 31st Annual Conference on Innovative Applications of Artificial Intelligence, IAAI 2019 and the 9th AAAI Symposium on Educational Advances in Artificial Intelligence, EAAI 2019
Y2 - 27 January 2019 through 1 February 2019
ER -