Natural Language Processing (NLP)

Undergraduate Study Group

Moderator: Mohammad Taher Pilehvar

Wednesdays, 10am-11am (online on Skyroom)

The aim of this study group is to lay the foundations for carrying out research. This includes studying the basics of NLP and Machine Learning (mostly Deep Learning), reading papers, and discussing ideas.

Contact the moderator if you are interested in joining the group.

Current members

Series 1

Summer 2020

Date Topic
30 Sep
  • Image captioning: implementing a network to generate captions for input images.
23 Sep
  • Review of the SQuAD dataset and Question Answering task.
  • Convolutional Neural Networks, a simple classification of MNIST.
16 Sep
  • Review of Collobert et al, Natural Language Processing (Almost) from Scratch, 2010.
  • Generative RNN models, Sequence to Sequence architecture. Application to a toy tasks, such as arithmetics, chatbot, decimal to binary conversion, etc.
9 Sep
2 Sep Review of two papers:
  • P. Turney, From Frequency to Meaning: Vector Space Models of Semantics, 2010.
  • Bolukbasi et al, Man is to Computer Programmer as Woman is to Homemaker, 2016.
26 Aug
  • Initiating Farsi NLP tracker, a collection of NLP datasets for Farsi.
  • Imbalanced classification
  • Precision/Recall/F1
19 Aug
  • Optimizers (presentation by Alireza Moradi)
  • NER task for Farsi on Persian-NER
  • TimeDistributed layer in Keras
12 Aug
  • Feeding pre-trained embeddings (fastText)
  • Freezing vs. tuning weights
  • More on recurrent neural networks (what is "unit" size?)
4 Aug --
29 July
  • Understanding the concept of semantic representation: Word embeddings (Word2vec)
  • Implementing a sentiment analysis model on DigiKala data, using LSTMs in Keras
22 July
15 July
8 July Introduction to the study group: aims and objectives.