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
440
4
Share
MixPoet
Zhang Yixiao
April 30, 2020
More Decks by Zhang Yixiao
See All by Zhang Yixiao
CoCon
ldzhangyx
0
380
vq-cpc
ldzhangyx
0
380
diora
ldzhangyx
0
290
drummernet
ldzhangyx
0
240
ON-LSTM
ldzhangyx
0
200
Other Decks in Research
See All in Research
ペットのかわいい瞬間を撮影する オートシャッターAIアプリへの スマートラベリングの適用
mssmkmr
0
500
RS-Agent: Automating Remote Sensing Tasks through Intelligent Agent
satai
0
240
はじまりの クエスチョンブック —余暇と豊かさにあふれた社会とは?
culturaltransition
PRO
0
470
R&Dチームを起ち上げる
shibuiwilliam
1
260
LOSの検討(λ Kansai 2026 in Winter)
motopu
0
130
SOTAのさらに先へ:厳しい推論制約下での高性能モデルのPost-Training
analokmaus
0
1.1k
Research Engineerという仕事 / Research Engineering: Bridging Research and Business
chck
1
170
【NICOGRAPH2025】Photographic Conviviality: ボディペイント・ワークショップによる 同時的かつ共生的な写真体験
toremolo72
0
240
社内データ分析AIエージェントを できるだけ使いやすくする工夫
fufufukakaka
1
1.1k
Using our influence and power for patient safety
helenbevan
0
350
Dual Quadric表現を用いた動的物体追跡とRGB-D・IMU制約の密結合によるオドメトリ推定
nanoshimarobot
0
390
The Landscape of Agentic Reinforcement Learning for LLMs: A Survey
shunk031
4
970
Featured
See All Featured
Responsive Adventures: Dirty Tricks From The Dark Corners of Front-End
smashingmag
254
22k
Raft: Consensus for Rubyists
vanstee
141
7.5k
How to Grow Your eCommerce with AI & Automation
katarinadahlin
PRO
1
190
Applied NLP in the Age of Generative AI
inesmontani
PRO
4
2.3k
We Analyzed 250 Million AI Search Results: Here's What I Found
joshbly
1
1.3k
The #1 spot is gone: here's how to win anyway
tamaranovitovic
2
1.1k
Why Mistakes Are the Best Teachers: Turning Failure into a Pathway for Growth
auna
0
150
StorybookのUI Testing Handbookを読んだ
zakiyama
31
6.8k
So, you think you're a good person
axbom
PRO
2
2k
A better future with KSS
kneath
240
18k
Learning to Love Humans: Emotional Interface Design
aarron
275
41k
How To Stay Up To Date on Web Technology
chriscoyier
790
250k
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