Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
MixPoet
Search
Sponsored
·
Your Podcast. Everywhere. Effortlessly.
Share. Educate. Inspire. Entertain. You do you. We'll handle the rest.
→
Zhang Yixiao
April 30, 2020
Research
460
4
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
MixPoet
Zhang Yixiao
April 30, 2020
More Decks by Zhang Yixiao
See All by Zhang Yixiao
CoCon
ldzhangyx
0
410
vq-cpc
ldzhangyx
0
390
diora
ldzhangyx
0
300
drummernet
ldzhangyx
0
260
ON-LSTM
ldzhangyx
0
230
Other Decks in Research
See All in Research
National high-resolution cropland classification of Japan with agricultural census information and multi-temporal multi-modality datasets
satai
3
410
計算情報学研究室(数理情報学第7研究室)2026
tomohirokoana
0
710
AY 2026 Guide to Academic Writing Using Generative AI - Workshop
ks91
PRO
0
150
MIRU2026 チュートリアル講演2:三次元データ処理の動向
nnchiba
3
2.9k
SLAMはどこまで解決されたのか?
tomonom
0
1k
PHTalks Bengaluru - SSRF When All Else Fails
dk999
0
920
ros2-perf-multihost: 分散システムにおける客観的なアーキテクチャ評価フレームワーク
takasehideki
0
170
Visual SLAM未来予測 / Future Prediction in Visual SLAM
koide3
1
820
260624_NLP-colloquium: Hubness
de9uch1
1
160
Unified Audio Source Separation (Defense Slides)
kohei_1979
1
640
データサイエンティストの就労意識~2015 → 2026 一般(個人)会員アンケートより
datascientistsociety
PRO
0
560
【中間報告】国会議員の立法・政策実務を支える環境を巡る現状と課題
polipoli
0
400
Featured
See All Featured
What’s in a name? Adding method to the madness
productmarketing
PRO
24
4.1k
Statistics for Hackers
jakevdp
799
230k
Creating an realtime collaboration tool: Agile Flush - .NET Oxford
marcduiker
35
2.5k
Accessibility Awareness
sabderemane
1
170
Typedesign – Prime Four
hannesfritz
42
3.1k
Design in an AI World
tapps
1
270
How to Think Like a Performance Engineer
csswizardry
28
2.7k
Performance Is Good for Brains [We Love Speed 2024]
tammyeverts
12
1.8k
We Have a Design System, Now What?
morganepeng
55
8.2k
Odyssey Design
rkendrick25
PRO
2
750
How to Align SEO within the Product Triangle To Get Buy-In & Support - #RIMC
aleyda
2
1.8k
CoffeeScript is Beautiful & I Never Want to Write Plain JavaScript Again
sstephenson
162
16k
Transcript
MixPoet: Diverse Poetry Generation via Learning Controllable Mixed Latent Space
ArXiv: 2003.06094v1 Presenter: Yixiao Zhang
Overview • Idea: 诗人经历、历史背景等 => 诗歌风格多样化 • Methods: • semi-supervised
VAE • disentangling latent space to sub-spaces • each sub-space corresponds to one factor conditioning • adversarial training
Introduction • 近年的研究,主要考虑语义连贯、主题相关 • 存在diversity的困扰 • diversity: • 主题间多样性:给定两个topic words,生成不同的诗歌
• 主题内多样性:给定一个topic word,生成不同的诗歌 • * 现有的模型倾向于记住常见pattern
Introduction • 生活经历、历史背景、文学流派 => 影响风格
Introduction • MixPoet: semi-supervised VAE • 将latent space分解为sub-spaces,与影响因子一一对应 • 训练阶段:模型预测无label诗歌的factors
• 测试阶段:指定factor的值,生成风格化的诗歌
Related Work • 诗歌生成模型 (RNNs, Memory Models, etc. ) •
多样性的先前研究: • MRL system: 强化学习,鼓励选用高TF-IDF的词汇 • USPG: 无监督最大化style vector和诗歌的mutual information
Related Work • VAE文本生成/诗歌生成 • Yang et. al, 2018b: 学习context-conditioned
latent variable • Hu et al. 2017: 对生成的诗歌进行对抗训练,增强topic相关性 • CVAE 对话多样性: Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders, ACL 2017 • 本文的对抗:在latent space上做对抗训练
Method • topic keyword: mixture empirical distributions: labeled/ unlabeled
Method: Generator • GRU based model • 是length embedding
Method: Semi-supervised C-VAE • 目的是学习 • 引入z • 由于style与semantics耦合 •
不假设y与z的独立性,而是: • 顺序: w => y => z => x (无y label时)
Method: Semi-supervised C-VAE • then for labeled data: • 估计先验
• 和后验 分别使用一个network计算, recon时最小化KL散度。
Method: Semi-supervised C-VAE • labeled data is too limited •
将y看作另一个latent variable • 估计先验 • 和后验 分别使用一个MLP network计算, recon y时最小化KL散度。
Method: Semi-supervised C-VAE • Total Loss:
Method: Latent Space Mixture • 多个factor时的情形: • 独立性假设:
Method: Latent Space Mixture • How to learn mixed latent
space? • For Isotropic Gaussian Space:
Method: Latent Space Mixture • How to learn mixed latent
space? • For Universal Space: 对于condition: ita是噪声,delta是脉冲函数,c是w, y => 从分布中sample出一个值
Method: Latent Space Mixture • 之后使得discriminator区分这两个z • 估计KL散度: • 其中
就是discriminator
Experiments • factors: • 军旅生涯, 乡村生活, 其他 • 时代繁荣, 时代衰落
• => 6种style
Experiments • Baseline: • Ground Truth • C-VAE • USPG
• MRL: SOTA • fBasic, 监督学习模型
Experiments • 多样性,使用Jaccard Similarity指数评价,越低越好 • 诗歌质量:使用Language Model Score(LMS)评价 • 观察:
• 大多数模型倾向生成重复的短语 • MRL与Basic在intra部分只能生成极其相似的诗歌 • C-VAE情况类似
Experiments • Factor Control Results: • 测试生成的诗歌是否与给定因子类别一致
Experiments • 主观实验
Analysis: Style Mixture
Analysis