arxivnotes's People
arxivnotes's Issues
🚧 【MaliGAN】Maximum-Likelihood Augmented Discrete Generative Adversarial Networks
ひとこと
Gに返すrewardを対数尤度に対応するものに変えたもの?
リンク
https://arxiv.org/abs/1702.07983
投稿日付(yyyy-mm-dd)
2017-02-26
概要
新規性・差分
手法
結果
Chinese poem generation taskにおいてseqGANよりBLEUスコアとPerplexityにおいて良い結果を出した
コメント
🚧 【SeqGAN】SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient
リンク
https://arxiv.org/abs/1609.05473
ひとこと
GANに強化学習を用いて生成器が離散系列を学習することを可能にした
参考になる解説ページなど
https://www.slideshare.net/DeepLearningJP2016/dlseqgan-sequence-generative-adversarial-nets-with-policy-gradient
https://qiita.com/everylittle/items/19c4988a135d36150dc0
【まとめ】GANによる自然言語生成に関する論文まとめ
投稿日付 | 名称 | 論文タイトル | リンク | issue | ひとこと |
---|---|---|---|---|---|
2016-09-18 | SeqGAN | SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient | https://arxiv.org/abs/1609.05473 | #2 | 強化学習 |
2017-02-26 | MaliGAN | Maximum-Likelihood Augmented Discrete Generative Adversarial Networks | https://arxiv.org/abs/1702.07983 | ||
2017-05-31 | RankGAN | Adversarial Ranking for Language Generation | https://arxiv.org/pdf/1705.11001 | ||
2017-06-12 | TextGAN | Adversarial Feature Matching for Text Generation | http://proceedings.mlr.press/v70/zhang17b.html | ||
2017-09-24 | LeakGAN | Long Text Generation via Adversarial Training with Leaked Information | https://arxiv.org/abs/1709.08624 | ||
2018 | SentiGAN | SentiGAN: Generating Sentimental Texts via Mixture Adversarial Networks | https://www.ijcai.org/proceedings/2018/618 | ||
2018-01-23 | MaskGAN | MaskGAN: Better Text Generation via Filling in the______ | https://arxiv.org/abs/1801.07736 | ||
2018-02-05 | DPGAN | DP-GAN: Diversity-Promoting Generative Adversarial Network for Generating Informative and Diversified Text | https://arxiv.org/abs/1802.01345 | ||
2018-09-18 | RelGAN | RelGAN: Relational Generative Adversarial Networks for Text Generation | https://openreview.net/forum?id=rJedV3R5tm | ||
2018-10-11 | LaTextGAN | Adversarial Text Generation Without Reinforcement Learning | https://arxiv.org/abs/1810.06640 | ||
2018-12-16 | Tet-GAN | TET-GAN: Text Effects Transfer via Stylization and Destylization | https://arxiv.org/abs/1812.06384 | よく見たらNLGではない | |
2019 | JSDGAN | Adversarial Discrete Sequence Generation without Explicit NeuralNetworks as Discriminators | http://proceedings.mlr.press/v89/li19g.html | ||
2019-08-24 | DGGAN | DGSAN: Discrete Generative Self-Adversarial Network | https://arxiv.org/abs/1908.09127 | ||
2019-11-15 | CatGAN | CatGAN: Category-aware Generative Adversarial Networks with Hierarchical Evolutionary Learning for Category Text Generation | https://arxiv.org/abs/1911.06641 |
その他surveyなど
投稿日付 | タイトル | リンク |
---|---|---|
2018-05-15 | Neural Text Generation: Past, Present and Beyond | https://arxiv.org/abs/1803.07133 |
2018-12-06 | Language GANs Falling Short | https://arxiv.org/abs/1811.02549 |
2018-02-06 | Texygen: A Benchmarking Platform for Text Generation Models | https://arxiv.org/abs/1802.01886 |
🚧 【TextGAN】Adversarial Feature Matching for Text Generation
ひとこと
Soft-argmax approximationを用いてdiscreteな系列を学習・GANの目的関数を変更してmode collapsing problemを解消
リンク
http://proceedings.mlr.press/v70/zhang17b.html
投稿日付(yyyy-mm-dd)
2017-06-12
概要
新規性・差分
手法
GeneratorはLSTM、DiscriminatorはCNN
結果
BookCorpus+Arxivから50万文ずつの学習において、seqGANに比べてBLEUとKDEで良い結果を出した
コメント
🚧 【LeakGAN】Long Text Generation via Adversarial Training with Leaked Information
ひとこと
リンク
https://arxiv.org/abs/1709.08624
投稿日付(yyyy-mm-dd)
2017-09-24
概要
新規性・差分
手法
結果
EMNLP2017 WMT4(long-text), COCO Image Captions Dataset (middle-text), Chinese Poems(short-text)において、SeqGANとRankGANに比べてBLEUスコアが良かった。さらに、前2つについては人手によるチューリングテストも行い,seqGANより良い結果が出た。
コメント
🚧 【RankGAN】Adversarial Ranking for Language Generation
ひとこと
seqGANに加えて識別器をrankerに置き換え、新しい目的関数を導入した
リンク
https://arxiv.org/pdf/1705.11001
投稿日付(yyyy-mm-dd)
2017-05-31
概要
新規性・差分
手法
結果
Chinese poem generationにおいてBLEU-2とHuman score両方で、COCOデータセットとShakespeareのロミオとジュリエットにおいてSeqGANより良い結果を出した。
コメント
🚧 【CatGAN】CatGAN: Category-aware Generative Adversarial Networks with Hierarchical Evolutionary Learning for Category Text Generation
ひとこと
遺伝的アルゴリズムを使ってGeneratorを進化させる
リンク
https://arxiv.org/abs/1911.06641
投稿日付(yyyy-mm-dd)
2019-11-15
概要
新規性・差分
手法
結果
コメント
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.