סילבוס הקורס עיבוד שפה טבעית (הקורס ינתן בשפה האנגלית) - תש"ף, פקולטה למדעים מדויקים, אוניברסיטת ת"א

שנה"ל תש"ף

עיבוד שפה טבעית (הקורס ינתן בשפה האנגלית)
Natural Language Processing

0368-3077

מדעים מדויקים

קבוצה 01

סמ' א'	0900-1200	'ד	005	שיעור	פרופ ברנט יהונתן
הקורס מועבר באנגלית
D:\Inetpub\shared\yedion\syllabus\03\2019\0368\0368307701_desc.txt סילבוס מקוצר Natural Language Processing (NLP) aims to develop methods for processing, analyzing and understanding natural language. The goal of this class is to provide a thorough overview of modern methods in the field of Natural Language Processing. The class will not assume prior knowledge in NLP. We will cover: Word representations Language modeling (LSTMs, Transformers, etc.) Tagging (sequence models) Parsing (trees) sequence-to-sequence models Course description This course will provide an overview of modern natural language processing, focusing on tasks where the output is a structure. We will cover language models, sequence models, and tree parsing models, focusing on the interface between structured prediction and deep learning. מדעים מדויקים 0368-3077-01 עיבוד שפה טבעית (הקורס ינתן בשפה האנגלית) Natural Language Processing שנה"ל תש"ף \| סמ' א' \| פרופ ברנט יהונתן סילבוס מפורט/דף מידע Natural Language Processing (NLP) aims to develop methods for processing, analyzing and understanding natural language. The goal of this class is to provide a thorough overview of modern methods in the field of Natural Language Processing. The class will not assume prior knowledge in NLP. Syllabus: word embeddings (word2vec and friends) Language models n-gram langauge modles feed-forward LMs RNN LMs Vanishing and exploding gradients Sequence models LSTMs GRUs Transformers Contextualized word representations (BERT, etc.) Tagging HMMs Locally normalized log-linear models Globally-normalized log-linear models Viterbi Forward-backward Deep tagging models (BiLSTMs) Syntactic Parsing CFGs, PCFGs CKY Lexicalized PCFGs Local and global log-linear parsing models Deep syntactic parsing models Sequence-to-sequence models attention pointer networks semantic parsing Sub-word models

ש"ס: 3.0

סילבוס מקוצר

Natural Language Processing (NLP) aims to develop methods for processing, analyzing and understanding natural language. The goal of this class is to provide a thorough overview of modern methods in the field of Natural Language Processing. The class will not assume prior knowledge in NLP.

We will cover:
- Word representations
- Language modeling (LSTMs, Transformers, etc.)
- Tagging (sequence models)
- Parsing (trees)
- sequence-to-sequence models

Course description

This course will provide an overview of modern natural language processing, focusing on tasks where the output is a structure. We will cover language models, sequence models, and tree parsing models, focusing on the interface between structured prediction and deep learning.

מדעים מדויקים
0368-3077-01 עיבוד שפה טבעית (הקורס ינתן בשפה האנגלית)
Natural Language Processing
שנה"ל תש"ף | סמ' א' | פרופ ברנט יהונתן

666סילבוס מפורט/דף מידע

Syllabus:

word embeddings (word2vec and friends)
Language models
- n-gram langauge modles
- feed-forward LMs
- RNN LMs
- Vanishing and exploding gradients
Sequence models
- LSTMs
- GRUs
- Transformers
Contextualized word representations (BERT, etc.)
Tagging
- HMMs
- Locally normalized log-linear models
- Globally-normalized log-linear models
- Viterbi
- Forward-backward
- Deep tagging models (BiLSTMs)
Syntactic Parsing
- CFGs, PCFGs
- CKY
- Lexicalized PCFGs
- Local and global log-linear parsing models
- Deep syntactic parsing models
Sequence-to-sequence models
- attention
- pointer networks
- semantic parsing
Sub-word models

להצהרת הנגישות