# Index

* [소개글](/natural-language-processing-with-pytorch/master.md)
* [서문](/natural-language-processing-with-pytorch/preface.md)
* [Index](/natural-language-processing-with-pytorch/index_list.md)
* [딥러닝을 활용한 자연어 처리 개요](/natural-language-processing-with-pytorch/00-cover.md)
  * [자연어 처리란 무엇일까?](/natural-language-processing-with-pytorch/00-cover/01-intro.md)
  * [딥러닝 소개](/natural-language-processing-with-pytorch/00-cover/02-deeplearning.md)
  * [왜 자연어 처리는 어려울까?](/natural-language-processing-with-pytorch/00-cover/03-why-nlp-difficult.md)
  * [무엇이 한국어 자연어 처리를 더욱 어렵게 만들까?](/natural-language-processing-with-pytorch/00-cover/04-korean-is-hell.md)
  * [자연어 처리의 최근 추세](/natural-language-processing-with-pytorch/00-cover/05-trends.md)
* [기초 수학](/natural-language-processing-with-pytorch/00-cover-1.md)
  * [서문](/natural-language-processing-with-pytorch/00-cover-1/01-intro.md)
  * [랜덤 변수와 확률 분포](/natural-language-processing-with-pytorch/00-cover-1/02-prob-dist.md)
  * [쉬어가기: 몬티 홀 문제](/natural-language-processing-with-pytorch/00-cover-1/03-monty-hall.md)
  * [기대값과 샘플링](/natural-language-processing-with-pytorch/00-cover-1/04-sampling.md)
  * [Maximum Likelihood Estimation](/natural-language-processing-with-pytorch/00-cover-1/05-mle.md)
  * [정보 이론](/natural-language-processing-with-pytorch/00-cover-1/06-information.md)
  * [쉬어가기: MSE 손실 함수와 확률 분포 함수](/natural-language-processing-with-pytorch/00-cover-1/07-mse.md)
  * [마치며](/natural-language-processing-with-pytorch/00-cover-1/08-conclusion.md)
* [Hello 파이토치](/natural-language-processing-with-pytorch/00-cover-2.md)
  * [딥러닝을 시작하기 전에](/natural-language-processing-with-pytorch/00-cover-2/01-intro.md)
  * [설치 방법](/natural-language-processing-with-pytorch/00-cover-2/02-how-to-install.md)
  * [짧은 튜토리얼](/natural-language-processing-with-pytorch/00-cover-2/03-hello-pytorch.md)
  * [쉬어가기: 윈도우즈 개발 환경 구축](/natural-language-processing-with-pytorch/00-cover-2/04-how_to_install_wins.md)
* [전처리](/natural-language-processing-with-pytorch/00-cover-3.md)
  * [전처리란](/natural-language-processing-with-pytorch/00-cover-3/01-intro.md)
  * [코퍼스 수집](/natural-language-processing-with-pytorch/00-cover-3/02-collecting-corpus.md)
  * [코퍼스 정제](/natural-language-processing-with-pytorch/00-cover-3/03-cleaning-corpus.md)
  * [분절](/natural-language-processing-with-pytorch/00-cover-3/04-tokenization.md)
  * [병렬 코퍼스 정렬](/natural-language-processing-with-pytorch/00-cover-3/05-align.md)
  * [서브워드 분절](/natural-language-processing-with-pytorch/00-cover-3/06-bpe.md)
  * [분절 복원](/natural-language-processing-with-pytorch/00-cover-3/07-detokenization.md)
  * [토치텍스트](/natural-language-processing-with-pytorch/00-cover-3/08-torchtext.md)
* [유사성과 모호성](/natural-language-processing-with-pytorch/00-cover-4.md)
  * [단어의 의미](/natural-language-processing-with-pytorch/00-cover-4/01-intro.md)
  * [One-hot 인코딩](/natural-language-processing-with-pytorch/00-cover-4/02-one-hot-encoding.md)
  * [시소러스를 활용한 단어 의미 파악](/natural-language-processing-with-pytorch/00-cover-4/03-wordnet.md)
  * [특징](/natural-language-processing-with-pytorch/00-cover-4/04-feature.md)
  * [특징 추출하기: TF-IDF](/natural-language-processing-with-pytorch/00-cover-4/05-tf-idf.md)
  * [특징 벡터 만들기](/natural-language-processing-with-pytorch/00-cover-4/06-vectorization.md)
  * [특징 유사도 구하기](/natural-language-processing-with-pytorch/00-cover-4/07-similarity.md)
  * [단어 중의성 해소](/natural-language-processing-with-pytorch/00-cover-4/08-wsd.md)
  * [Selectional Preference](/natural-language-processing-with-pytorch/00-cover-4/09-selectional-preference.md)
  * [마치며](/natural-language-processing-with-pytorch/00-cover-4/10-conclusion.md)
* [단어 임베딩](/natural-language-processing-with-pytorch/00-cover-5.md)
  * [들어가며](/natural-language-processing-with-pytorch/00-cover-5/01-intro.md)
  * [차원 축소](/natural-language-processing-with-pytorch/00-cover-5/02-dimension-reduction.md)
  * [흔한 오해 1](/natural-language-processing-with-pytorch/00-cover-5/03-myth.md)
  * [Word2Vec](/natural-language-processing-with-pytorch/00-cover-5/04-word2vec.md)
  * [GloVe](/natural-language-processing-with-pytorch/00-cover-5/05-glove.md)
  * [Word2Vec 예제](/natural-language-processing-with-pytorch/00-cover-5/06-example.md)
  * [마치며](/natural-language-processing-with-pytorch/00-cover-5/07-conclusion.md)
* [시퀀스 모델링](/natural-language-processing-with-pytorch/00-cover-6.md)
  * [들어가며](/natural-language-processing-with-pytorch/00-cover-6/01-intro.md)
  * [Recurrent Neural Network](/natural-language-processing-with-pytorch/00-cover-6/02-rnn.md)
  * [Long Short Term Memory](/natural-language-processing-with-pytorch/00-cover-6/03-lstm.md)
  * [Gated Recurrent Unit](/natural-language-processing-with-pytorch/00-cover-6/04-gru.md)
  * [그래디언트 클리핑](/natural-language-processing-with-pytorch/00-cover-6/05-gradient-clipping.md)
  * [마치며](/natural-language-processing-with-pytorch/00-cover-6/06-conclusion.md)
* [텍스트 분류](/natural-language-processing-with-pytorch/00-cover-7.md)
  * [들어가기](/natural-language-processing-with-pytorch/00-cover-7/01-intro.md)
  * [나이브 베이즈를 활용하기](/natural-language-processing-with-pytorch/00-cover-7/02-naive-bayes.md)
  * [흔한 오해 2](/natural-language-processing-with-pytorch/00-cover-7/03-myth.md)
  * [RNN을 활용하기](/natural-language-processing-with-pytorch/00-cover-7/04-rnn.md)
  * [CNN을 활용하기](/natural-language-processing-with-pytorch/00-cover-7/05-cnn.md)
  * [쉬어가기: 멀티 레이블 분류](/natural-language-processing-with-pytorch/00-cover-7/06-multi_classification.md)
  * [마치며](/natural-language-processing-with-pytorch/00-cover-7/07-conclusion.md)
* [언어 모델링](/natural-language-processing-with-pytorch/00-cover-8.md)
  * [들어가며](/natural-language-processing-with-pytorch/00-cover-8/01-intro.md)
  * [n-gram](/natural-language-processing-with-pytorch/00-cover-8/02-n-gram.md)
  * [언어 모델의 평가 방법](/natural-language-processing-with-pytorch/00-cover-8/03-perpexity.md)
  * [SRILM을 활용한 n-gram 실습](/natural-language-processing-with-pytorch/00-cover-8/04-srilm.md)
  * [NNLM](/natural-language-processing-with-pytorch/00-cover-8/05-nnlm.md)
  * [언어 모델의 활용](/natural-language-processing-with-pytorch/00-cover-8/06-application.md)
  * [마치며](/natural-language-processing-with-pytorch/00-cover-8/07-conclusion.md)
* [신경망 기계번역](/natural-language-processing-with-pytorch/00-cover-9.md)
  * [들어가며](/natural-language-processing-with-pytorch/00-cover-9/01-intro.md)
  * [Sequence-to-Sequence](/natural-language-processing-with-pytorch/00-cover-9/02-seq2seq.md)
  * [Attention](/natural-language-processing-with-pytorch/00-cover-9/03-attention.md)
  * [Input Feeding](/natural-language-processing-with-pytorch/00-cover-9/04-input-feeding.md)
  * [자기회귀 속성과 Teacher Forcing 훈련 방법](/natural-language-processing-with-pytorch/00-cover-9/05-teacher-forcing.md)
  * [탐색(추론)](/natural-language-processing-with-pytorch/00-cover-9/06-beam-search.md)
  * [성능 평가](/natural-language-processing-with-pytorch/00-cover-9/07-eval.md)
  * [마치며](/natural-language-processing-with-pytorch/00-cover-9/08-conclusion.md)
* [신경망 기계번역 심화 주제](/natural-language-processing-with-pytorch/00-cover-10.md)
  * [다국어 신경망 번역](/natural-language-processing-with-pytorch/00-cover-10/01-multilingual-nmt.md)
  * [단일 언어 코퍼스를 활용하기](/natural-language-processing-with-pytorch/00-cover-10/02-monolingual-corpus.md)
  * [트랜스포머](/natural-language-processing-with-pytorch/00-cover-10/03-transformer.md)
  * [마치며](/natural-language-processing-with-pytorch/00-cover-10/04-conclusion.md)
* [강화학습을 활용한 자연어 생성](/natural-language-processing-with-pytorch/00-cover-11.md)
  * [들어가며](/natural-language-processing-with-pytorch/00-cover-11/01-intro.md)
  * [강화학습 기초](/natural-language-processing-with-pytorch/00-cover-11/02-rl_basics.md)
  * [정책 기반 강화학습](/natural-language-processing-with-pytorch/00-cover-11/03-policy-gradient.md)
  * [자연어 생성에 강화학습 적용하기](/natural-language-processing-with-pytorch/00-cover-11/04-characteristic.md)
  * [강화학습을 활용한 지도학습](/natural-language-processing-with-pytorch/00-cover-11/05-supervised-nmt.md)
  * [강화학습을 활용한 비지도학습](/natural-language-processing-with-pytorch/00-cover-11/06-unsupervised-nmt.md)
  * [마치며](/natural-language-processing-with-pytorch/00-cover-11/07-conclusion.md)
* [듀얼리티 활용](/natural-language-processing-with-pytorch/00-cover-12.md)
  * [들어가며](/natural-language-processing-with-pytorch/00-cover-12/01-intro.md)
  * [듀얼리티를 활용한 지도학습](/natural-language-processing-with-pytorch/00-cover-12/02-dsl.md)
  * [듀얼리티를 활용한 비지도학습](/natural-language-processing-with-pytorch/00-cover-12/03-dul.md)
  * [쉬어가기: Back-translation을 재해석하기](/natural-language-processing-with-pytorch/00-cover-12/04-back_translation.md)
  * [마치며](/natural-language-processing-with-pytorch/00-cover-12/05-conclusion.md)
* [NMT 시스템 구축](/natural-language-processing-with-pytorch/00-cover-13.md)
  * [파이프라인](/natural-language-processing-with-pytorch/00-cover-13/01-pipeline.md)
  * [구글의 NMT](/natural-language-processing-with-pytorch/00-cover-13/02-gnmt.md)
  * [에딘버러 대학의 NMT](/natural-language-processing-with-pytorch/00-cover-13/03-nematus.md)
  * [MS의 NMT](/natural-language-processing-with-pytorch/00-cover-13/04-microsoft.md)
* [전이학습](/natural-language-processing-with-pytorch/00-cover-14.md)
  * [전이학습이란?](/natural-language-processing-with-pytorch/00-cover-14/01-intro.md)
  * [기존의 사전 훈련 방식](/natural-language-processing-with-pytorch/00-cover-14/02-previous_work.md)
  * [ELMo](/natural-language-processing-with-pytorch/00-cover-14/03-elmo.md)
  * [BERT](/natural-language-processing-with-pytorch/00-cover-14/04-bert.md)
  * [GPT-2](/natural-language-processing-with-pytorch/00-cover-14/05-gpt.md)
  * [XLNet](/natural-language-processing-with-pytorch/00-cover-14/06-xlnet.md)
  * [마치며](/natural-language-processing-with-pytorch/00-cover-14/07-conclusion.md)
* [이 책을 마치며](/natural-language-processing-with-pytorch/epilogue.md)
* [참고문헌](/natural-language-processing-with-pytorch/references.md)


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://kh-kim.gitbook.io/natural-language-processing-with-pytorch/index_list.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.