@inproceedings{yu-etal-2023-probabilistic,
title = "Probabilistic Robustness for Data Filtering",
author = "Yu, Yu and
Khan, Abdul Rafae and
Khadivi, Shahram and
Xu, Jia",
editor = "Vlachos, Andreas and
Augenstein, Isabelle",
booktitle = "Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics",
month = may,
year = "2023",
address = "Dubrovnik, Croatia",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.eacl-main.214",
doi = "10.18653/v1/2023.eacl-main.214",
pages = "2950--2959",
abstract = "We introduce our probabilistic robustness rewarded data optimization (PRoDO) approach as a framework to enhance the model{'}s generalization power by selecting training data that optimizes our probabilistic robustness metrics. We use proximal policy optimization (PPO) reinforcement learning to approximately solve the computationally intractable training subset selection problem. The PPO{'}s reward is defined as our ($\alpha,\epsilon, \gamma$)-Robustness that measures performance consistency over multiple domains by simulating unknown test sets in real-world scenarios using a leaving-one-out strategy. We demonstrate that our PRoDO effectively filters data that lead to significantly higher prediction accuracy and robustness on unknown-domain test sets. Our experiments achieve up to +17.2{\%} increase of accuracy (+25.5{\%} relatively) in sentiment analysis, and -28.05 decrease of perplexity (-32.1{\%} relatively) in language modeling.In addition, our probabilistic ($\alpha,\epsilon, \gamma$)-Robustness definition serves as an evaluation metric with higher levels of agreement with human annotations than typical performance-based metrics.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yu-etal-2023-probabilistic">
<titleInfo>
<title>Probabilistic Robustness for Data Filtering</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yu</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Abdul</namePart>
<namePart type="given">Rafae</namePart>
<namePart type="family">Khan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shahram</namePart>
<namePart type="family">Khadivi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jia</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Andreas</namePart>
<namePart type="family">Vlachos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Isabelle</namePart>
<namePart type="family">Augenstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dubrovnik, Croatia</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We introduce our probabilistic robustness rewarded data optimization (PRoDO) approach as a framework to enhance the model’s generalization power by selecting training data that optimizes our probabilistic robustness metrics. We use proximal policy optimization (PPO) reinforcement learning to approximately solve the computationally intractable training subset selection problem. The PPO’s reward is defined as our (α,ε, γ)-Robustness that measures performance consistency over multiple domains by simulating unknown test sets in real-world scenarios using a leaving-one-out strategy. We demonstrate that our PRoDO effectively filters data that lead to significantly higher prediction accuracy and robustness on unknown-domain test sets. Our experiments achieve up to +17.2% increase of accuracy (+25.5% relatively) in sentiment analysis, and -28.05 decrease of perplexity (-32.1% relatively) in language modeling.In addition, our probabilistic (α,ε, γ)-Robustness definition serves as an evaluation metric with higher levels of agreement with human annotations than typical performance-based metrics.</abstract>
<identifier type="citekey">yu-etal-2023-probabilistic</identifier>
<identifier type="doi">10.18653/v1/2023.eacl-main.214</identifier>
<location>
<url>https://aclanthology.org/2023.eacl-main.214</url>
</location>
<part>
<date>2023-05</date>
<extent unit="page">
<start>2950</start>
<end>2959</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Probabilistic Robustness for Data Filtering
%A Yu, Yu
%A Khan, Abdul Rafae
%A Khadivi, Shahram
%A Xu, Jia
%Y Vlachos, Andreas
%Y Augenstein, Isabelle
%S Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics
%D 2023
%8 May
%I Association for Computational Linguistics
%C Dubrovnik, Croatia
%F yu-etal-2023-probabilistic
%X We introduce our probabilistic robustness rewarded data optimization (PRoDO) approach as a framework to enhance the model’s generalization power by selecting training data that optimizes our probabilistic robustness metrics. We use proximal policy optimization (PPO) reinforcement learning to approximately solve the computationally intractable training subset selection problem. The PPO’s reward is defined as our (α,ε, γ)-Robustness that measures performance consistency over multiple domains by simulating unknown test sets in real-world scenarios using a leaving-one-out strategy. We demonstrate that our PRoDO effectively filters data that lead to significantly higher prediction accuracy and robustness on unknown-domain test sets. Our experiments achieve up to +17.2% increase of accuracy (+25.5% relatively) in sentiment analysis, and -28.05 decrease of perplexity (-32.1% relatively) in language modeling.In addition, our probabilistic (α,ε, γ)-Robustness definition serves as an evaluation metric with higher levels of agreement with human annotations than typical performance-based metrics.
%R 10.18653/v1/2023.eacl-main.214
%U https://aclanthology.org/2023.eacl-main.214
%U https://doi.org/10.18653/v1/2023.eacl-main.214
%P 2950-2959
Markdown (Informal)
[Probabilistic Robustness for Data Filtering](https://aclanthology.org/2023.eacl-main.214) (Yu et al., EACL 2023)
ACL
- Yu Yu, Abdul Rafae Khan, Shahram Khadivi, and Jia Xu. 2023. Probabilistic Robustness for Data Filtering. In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, pages 2950–2959, Dubrovnik, Croatia. Association for Computational Linguistics.