Global Autoregressive Models for Data-Efficient Sequence Learning

Tetiana Parshakova, Jean-Marc Andreoli, Marc Dymetman


Abstract
Standard autoregressive seq2seq models are easily trained by max-likelihood, but tend to show poor results under small-data conditions. We introduce a class of seq2seq models, GAMs (Global Autoregressive Models), which combine an autoregressive component with a log-linear component, allowing the use of global a priori features to compensate for lack of data. We train these models in two steps. In the first step, we obtain an unnormalized GAM that maximizes the likelihood of the data, but is improper for fast inference or evaluation. In the second step, we use this GAM to train (by distillation) a second autoregressive model that approximates the normalized distribution associated with the GAM, and can be used for fast inference and evaluation. Our experiments focus on language modelling under synthetic conditions and show a strong perplexity reduction of using the second autoregressive model over the standard one.
Anthology ID:
K19-1084
Volume:
Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL)
Month:
November
Year:
2019
Address:
Hong Kong, China
Venue:
CoNLL
SIG:
SIGNLL
Publisher:
Association for Computational Linguistics
Note:
Pages:
900–909
Language:
URL:
https://www.aclweb.org/anthology/K19-1084
DOI:
10.18653/v1/K19-1084
Bib Export formats:
BibTeX MODS XML EndNote
PDF:
http://aclanthology.lst.uni-saarland.de/K19-1084.pdf
Attachment:
 K19-1084.Attachment.pdf
Supplementary material:
 K19-1084.Supplementary_Material.pdf