Diferencias

Muestra las diferencias entre dos versiones de la página.

--- materias:pln:uba2019:teoricos [2019/09/10 22:35] – francolq
+++ materias:pln:uba2019:teoricos [2019/09/23 18:09] (actual) – francolq
@@ Línea 35: / Línea 35: @@
     * [[https://www.youtube.com/watch?v=ZbHFLgBWgdQ&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm&index=16|Suavizado "add one"]]
     * [[https://www.youtube.com/watch?v=naNezonMA7k&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm&index=17|Interpolación]]
+  * Notebooks:
+    * [[https://github.com/PLN-FaMAF/pln-uba-2019/blob/master/notebooks/01%20Procesamiento%20B%C3%A1sico%20de%20Texto.ipynb|01 Procesamiento Básico de Texto]]
+    * [[https://github.com/PLN-FaMAF/pln-uba-2019/blob/master/notebooks/02%20Modelado%20de%20Lenguaje.ipynb|02 Modelado de Lenguaje]]
   * Material complementario:
@@ Línea 42: / Línea 46: @@
+===== 2da clase =====
+  * Etiquetado de secuencias:
+    * [[https://www.youtube.com/watch?v=JhJU0Akkqzo&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm&index=56|Introducción al Part-of-Speech (POS) tagging]]
+    * [[https://www.youtube.com/watch?v=Zm_bmRhbaQg&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm&index=57|Algunos métodos y resultados]]
+  * Más sobre etiquetado de secuencias (curso de Collins):
+    * [[https://www.youtube.com/watch?v=6jUva-eD-xY&list=PLlQBy7xY8mbI13gwXZz4r55MeatSZOqm7&index=1|Etiquetado de secuencias]]
+    * [[https://www.youtube.com/watch?v=VMZM7AYjEsg&list=PLlQBy7xY8mbI13gwXZz4r55MeatSZOqm7&index=2|Modelos generativos para el aprendizaje supervisado]]
+    * [[https://www.youtube.com/watch?v=cAJmM5k62yM&list=PLlQBy7xY8mbI13gwXZz4r55MeatSZOqm7&index=3|Introducción a los Hidden Markov Models (HMMs)]]
+    * [[https://www.youtube.com/watch?v=uAT3iJpQwJ0&list=PLlQBy7xY8mbI13gwXZz4r55MeatSZOqm7&index=4|Estimación de parámetros de HMMs]]
+    * [[https://www.youtube.com/watch?v=ECu_KQV3V30&list=PLlQBy7xY8mbI13gwXZz4r55MeatSZOqm7&index=5|5. The Viterbi Algorithm for HMMs - Part I]]
+    * [[https://www.youtube.com/watch?v=WqGUa54x8wE&list=PLlQBy7xY8mbI13gwXZz4r55MeatSZOqm7&index=6|6. The Viterbi Algorithm for HMMs - Part II]]
+    * [[https://www.youtube.com/watch?v=Bu7oSlNCmdU&list=PLlQBy7xY8mbI13gwXZz4r55MeatSZOqm7&index=7|7. The Viterbi Algorithm for HMMs - Part III]]
+    * [[https://www.youtube.com/watch?v=Y5hXE23Tdzc&list=PLlQBy7xY8mbI13gwXZz4r55MeatSZOqm7&index=8|8. Summary]]
+  * Notebooks:
+    * [[https://github.com/PLN-FaMAF/pln-uba-2019/blob/master/notebooks/tagging/01%20Etiquetado%20de%20Secuencias%20Parte%201.ipynb|01 Etiquetado de Secuencias Parte 1]]
+    * [[https://github.com/PLN-FaMAF/pln-uba-2019/blob/master/notebooks/tagging/02%20Etiquetado%20de%20Secuencias%20Parte%202.ipynb|02 Etiquetado de Secuencias Parte 2]]
+    * [[https://github.com/PLN-FaMAF/pln-uba-2019/blob/master/notebooks/tagging/03%20Etiquetado%20de%20Secuencias%20Parte%203.ipynb|03 Etiquetado de Secuencias Parte 3]]
+{{:materias:pln:2019:errorsmeme.png?direct&400|}}
+  * Clasificación de texto
+    * [[https://www.youtube.com/watch?v=kxImnFg4ZiQ&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm&index=24|Qué es clasificación de texto]]
+    * [[https://www.youtube.com/watch?v=j39c7Gjx2gE&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm&index=25|Naive Bayes]]
+    * [[https://www.youtube.com/watch?v=VNEdufXVMaU&index=26&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm|Formalizando el clasificador Naive Bayes]]
+    * [[https://www.youtube.com/watch?v=3jR8TZG8T88&index=27&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm|Aprendizaje de Naive Bayes]]
+    * [[https://www.youtube.com/watch?v=LRFdF9J__Tc&index=28&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm|Relación de Naive Bayes con modelado de lenguaje]]
+    * [[https://www.youtube.com/watch?v=OWGVQfuvNMk&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm&index=29|Multinomial Naive Bayes: Ejemplo completo]]
+    * [[https://www.youtube.com/watch?v=81j2nzzBHUw&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm&index=30|Precision, recall y F1]]
+    * [[https://www.youtube.com/watch?v=TdkWIxGoiak&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm&index=31|Evaluación de clasificación de texto]]
+  * Análisis de Sentimiento:
+    * [[https://www.youtube.com/watch?v=vy0HC5H-484&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm&index=33|What is Sentiment Analysis]]
+    * [[https://www.youtube.com/watch?v=Dgqt62RQMaY&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm&index=34|Sentiment Analysis-  A baseline algorithm]]
+    * [[https://www.youtube.com/watch?v=wBE0FE_2ddE&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm&index=35|Sentiment Lexicons]]
+    * [[https://www.youtube.com/watch?v=Z7RxBcpyN1U&list=PLQiyVNMpDLKnZYBTUOlSI9mi9wAErFtFm&index=36|Learning Sentiment Lexicons]]
+===== 3ra clase =====
+  * Estrategias para Machine Learning:
+    * [[https://docs.google.com/presentation/d/e/2PACX-1vSjH0TlJzJpY3JeWWY_vPQpHUQnOzcg9cEMLzAcXj8cnm00l8G9_2a9L8eyB6aDWlpUgS0dOTE88j4y/pub?start=false&loop=false&delayms=3000|Parte 1]]
+    * [[https://docs.google.com/presentation/d/e/2PACX-1vQkXr831rL-O4iqzsWN1a7vqGoXSww-5qfdHomFu5AF5_hWC9QBo984Il92jbQ3LLvQmF73Pksib13m/pub?start=false&loop=false&delayms=3000|Parte 2]]
+  * Notebooks:
+    * [[https://github.com/PLN-FaMAF/pln-uba-2019/blob/master/notebooks/sentiment/01%20Baseline.ipynb|01 Baseline]]
+    * [[https://github.com/PLN-FaMAF/pln-uba-2019/blob/master/notebooks/sentiment/02%20Bag%20of%20Words.ipynb|02 Bag of Words]]
+    * [[https://github.com/PLN-FaMAF/pln-uba-2019/blob/master/notebooks/sentiment/03%20Clasificador%20Basico.ipynb|03 Clasificador Basico]]
+    * [[https://github.com/PLN-FaMAF/pln-uba-2019/blob/master/notebooks/sentiment/04%20Modelos%20de%20Clasificacion.ipynb|04 Modelos de Clasificacion]]
+    * [[https://github.com/PLN-FaMAF/pln-uba-2019/tree/master/notebooks/sentiment|Más...]]
+  * Charla de Rodrigo Loredo en el marco del seminario LIIA.
+  * Material complementario:
+    * [[https://www.deeplearning.ai/machine-learning-yearning/|Andrew Ng. “Machine Learning Yearning”. Draft, 2018.]]
+    * [[https://karpathy.github.io/2019/04/25/recipe/|A Recipe for Training Neural Networks (Andrej Karpathy)]]
+    * [[https://github.com/EpistasisLab/tpot|TPOT: A Python Automated Machine Learning tool that optimizes machine learning pipelines using genetic programming.]]
+===== 4ta clase =====
+/*
+    * [[http://web.stanford.edu/class/cs224n/lectures/lecture1.pdf|Introduction to NLP and Deep Learning]] ([[https://cs.famaf.unc.edu.ar/~francolq/uba2018/lecture1-2.pdf|versión corta]], [[https://www.youtube.com/watch?v=OQQ-W_63UgQ&t=2725s&index=1&list=PL3FW7Lu3i5Jsnh1rnUwq_TcylNr7EkRe6|videolecture]])
+*/
+  * Word Vectors:
+    * [[http://web.stanford.edu/class/cs224n/slides/cs224n-2019-lecture01-wordvecs1.pdf|Word Vectors 1]] ([[https://youtu.be/8rXD5-xhemo|videolecture]])
+    * [[http://web.stanford.edu/class/cs224n/slides/cs224n-2019-lecture02-wordvecs2.pdf|Word Vectors 2]] ([[https://youtu.be/kEMJRjEdNzM|videolecture]])
+  * Material complementario:
+    * [[https://docs.google.com/document/d/18NoNdArdzDLJFQGBMVMsQ-iLOowP1XXDaSVRmYN0IyM/edit|Frontiers in Natural Language Processing Expert Responses:]] encuesta a referentes del área. En particular:
+      * What would you say is the most influential work in NLP in the last decade, if you had to pick just one?
+===== 5ta clase =====
+  * Introducción a Redes Neuronales:
+    * [[https://www.youtube.com/watch?v=8CWyBNX6eDo&list=PLoROMvodv4rOhcuXMZkNm7j3fVwBBY42z&index=3|Neural Networks (cs224n lecture 3)]]
+    * [[https://www.youtube.com/watch?v=yLYHDSv-288&list=PLoROMvodv4rOhcuXMZkNm7j3fVwBBY42z&index=4|Backpropagation (cs224n lecture 4)]]
+  * Análisis sintáctico:
+    * [[https://youtu.be/nC9_RfjYwqA|Linguistic Structure: Dependency Parsing (cs224n lecture 5)]]
+  * Redes Neuronales Recurrentes:
+    * [[https://www.youtube.com/watch?v=iWea12EAu6U&list=PLoROMvodv4rOhcuXMZkNm7j3fVwBBY42z&index=6|Language Models and RNNs (cs224n lecture 6)]]
+    * [[https://www.youtube.com/watch?v=QEw0qEa0E50&list=PLoROMvodv4rOhcuXMZkNm7j3fVwBBY42z&index=7|Vanishing Gradients, Fancy RNNs (cs224n lecture 7)]]
+  * Links:
+    * [[http://colah.github.io/posts/2015-08-Understanding-LSTMs/|Understanding LSTM Networks]]
+===== 6ta clase =====
+{{:materias:pln:2019:deepmeme.png?direct&400|}}
+  * Traducción Automática y modelos "sequence to sequence":
+    * [[https://www.youtube.com/watch?v=XXtpJxZBa2c&list=PLoROMvodv4rOhcuXMZkNm7j3fVwBBY42z&index=8| Translation, Seq2Seq, Attention (cs224n lecture 8)]]
+  * [[https://www.youtube.com/watch?v=yIdF-17HwSk&list=PLoROMvodv4rOhcuXMZkNm7j3fVwBBY42z&index=11|Question Answering (cs224n lecture 10)]]
+  * [[https://www.youtube.com/watch?v=S-CspeZ8FHc&list=PLoROMvodv4rOhcuXMZkNm7j3fVwBBY42z&index=14|Contextual Word Embeddings (cs224n lecture 13)]]
+  * Material complementario:
+    * [[http://ruder.io/deep-learning-nlp-best-practices/index.html|Deep Learning for NLP Best Practices
+]] (Sebastian Ruder)
+    * [[http://nlp.seas.harvard.edu/2018/04/03/attention.html|The Annotated Transformer]] (Alexander Rush)
+    * [[https://talktotransformer.com/]]
+    * [[http://ruder.io/4-biggest-open-problems-in-nlp/|The 4 Biggest Open Problems in NLP
+]] (Sebastian Ruder)