All things you want to know about Raphael Shu

My Name and Status

	Raphael Shu / 朱 中元 The University of Tokyo Nakayama Lab PhD Candidate

Research Interests

• Natural Language Processing
  • Natural Language Understanding
  • Machine Translation
• Deep Learning
  • Learning Discrete Latent Structures
  • Generative Models

Cool Stuffs

• Deep Learning Monitor (deeplearn.org)
• tree2code
  • Learning discrete codes for syntax
• lanmt
  • Non-autoregressive neural machine translation with latent variables and refinement
• NLPer Blog
• Visualization of optimization methods (Chrome-only)
• Visualization of Monte Carlo methods (Chrome-only)

My Papers

• Raphael Shu, Jason Lee, Hideki Nakayama, Kyunghyun Cho, “Latent-Variable Non-Autoregressive Neural Machine Translation with Deterministic Inference using a Delta Posterior”, Arxiv abs/1908.07181
  • [PDF] https://arxiv.org/abs/1908.07181
  • [Code] https://github.com/zomux/lanmt
• Raphael Shu, Hideki Nakayama and Kyunghyun Cho, “Generating Diverse Translations with Sentence Codes”, ACL 2019
  • [PDF] https://www.aclweb.org/anthology/P19-1177/
  • [Code] https://github.com/zomux/tree2code
• Jiali Yao, Raphael Shu, Xinjian Li, Katsutoshi Ohtsuki, Hideki Nakayama. “Enabling Real-time Neural IME with Incremental Vocabulary Selection”, NAACL 2019
  • [PDF] to be added
• 朱中元, 中山英樹, 「Generating Syntactically Diverse Translations with Syntactic Codes」, 言語処理学会 2019
  • Best Paper / 最優秀賞
    　　- Official Report
• Raphael Shu, Hideki Nakayama, “Improving Beam Search by Removing Monotonic Constraint for Neural Machine Translation”, ACL 2018
  • PDF
• 朱中元, 中山英樹, 「深層コード学習による単語分散表現の圧縮」, 言語処理学会 2018
  • Best Paper / 最優秀賞
  • PDF
  • code
• Raphael Shu, Hideki Nakayama, “Compressing Word Embeddings via Deep Compositional Code Learning”, ICLR 2018
  • PDF
  • code
• Raphael Shu, Hideki Nakayama, “An Empirical Study of Adequate Vision Span for Attention-Based Neural Machine Translation”, ACL Workshop on Neural Machine Translation, 2017
  • Outstanding Paper Award
  • http://aclweb.org/anthology/W/W17/W17-3201.pdf
• 朱中元, 中山英樹, 「文脈を考慮したアテンションメカニズムの計算量の削減」, 人工知能大会 2017
  • Student Paper Award / 学生奨励賞
  • PDF
• Evaluating Neural Machine Translation in English-Japanese Task
  • Second place in human evaluation, third place in JPO accuracy test
    • http://www.aclweb.org/anthology/W/W15/W15-5007.pdf
    • poster
• Weblio Pre-reordering Statistical Machine Translation System
  • Second place in English-Japanese translation task of WAT (Workshop on Asian Translation) 2014
    • paper
    • slides
    • poster
• A preordering method using head-restructured CFG parse tree for SMT
  • http://www.anlp.jp/proceedings/annual_meeting/2014/pdf_dir/P6-5.pdf
• TSUKU Statistical Machine Translation System for the NTCIR-10 PatentMT Task
  • http://research.nii.ac.jp/ntcir/workshop/OnlineProceedings10/pdf/NTCIR/PatentMT/20-NTCIR10-PATENTMT-ZhuZ.pdf
• Thesis of master’s degree
  • /files/master-thesis.pdf

My Codes

• tree2code: Learning Discrete Syntactic Codes for Structural Diverse Translation
  • https://github.com/zomux/tree2code
• nmtlab: A simplified framework for neural machine translation
  • https://github.com/zomux/nmtlab
• neuralcompressor
  • Compress word embeddings
  • https://github.com/zomux/neuralcompressor
• deepy
  • Highly extensible deep learning framework based on Theano
  • https://github.com/zomux/deepy
• neuralmt
  • Neural machine translation modelling tookit
  • https://github.com/zomux/neuralmt
• deepy-draw
  • An implementation of DRAW with deepy framework
  • 
  • https://github.com/zomux/deepy-draw/
• JRNNLM
  • RNNLM implementation in JAVA
  • https://github.com/zomux/jrnnlm

My Experiences

year	place	what done
2016 - 2019	The University of Tokyo, PhD Program	Neural Machine Translation / Multimodal Deep Learning / Discrete Structural Learning
2014 - 2016	Research & Development at Weblio, Inc.	Machine Translation / Conversation Models / Data Scientist
2012 - 2014	University of Tsukuba, Master Program	Machine Translation

How to find me

raphael [at] uaca [dot] com