Deep Learning in NLP - Powered by Python

Rishabh (~rishy)


19

Votes

Description:

Deep Learning has recently gained lot of attention in academia as well as in industry. Although the most primitive concepts of artificial intelligence date back to 1960s-1970s, but with the advent of more powerful machines and even more elegant programming languages(of course Python), now it's possible to code a Deep Learning model in a very short time and with less hassle.

Natural Language Processing(NLP) has been a very important sub-field of AI and current state-of-the-art models employs deep neural networks, most of the time coupled with Python to quickly develop these rather complex networks. Some of the use cases of NLP is Language Modelling, Sentiment Analysis, Named Entity Recognition, Topic Modelling, Semantic Proximity, etc. In this workshop we'll create a Deep Learning Model(Recurrent Neural Network) for Language Modelling using Gensim Word2Vec and Theano. Gensim and Theano are both Python based libraries which makes it real easy to code Deep Neural Nets.


Talk Structure


  • Introduction to Neural Networks
  • Word Vector Embeddings
  • Introduction to Gensim
  • Introduction to Theano
  • What are Deep Learning Models?
  • Recurrent Neural Networks
  • What is a Language Model?
  • A Language model with RNN and Theano

Prerequisites:

  • Basic Python and Numpy
  • Knowledge of Neural Networks


Required Setup for the Workshop


  1. Install "Conda" as suggested at: https://www.continuum.io/downloads
  2. conda install theano
  3. conda install gensim
  4. Download this pre-cleaned wikipedia articles file: Google Drive Link


Resources



Speaker Info:

I am the Platform Head and Chief Deep Learning Engineer at Neuron, a Natural Language Processing/Machine Learning startup. I am also a FOSS supporter and have been a GSOC Intern for Mifos Initiative in the year 2014.


Contact Info


Email - [email protected]
Phone - +917838742284

Section: Scientific Computing
Type: Workshops
Target Audience: Intermediate
Last Updated:

Can we have some links to your slides or a general structure, some thing that can be used to be put on the projector so audience can follow along.

Please upload the slides/structure so they can be reviewed before 12th feb.

Have you given any talks(including this one) before? Any experience of public speaking? It's not a requirement for doing the talk but would definitely help us gauge the experience level. We suggest going through the presentation least once in front of a small audience to get some experience if you have not already.

Akshay Arora (~akshayaurora)

Please find the bare bone structure of slides here: http://slides.com/rishabhshukla/deep-learning-in-nlp/live#/. I'll further add links for Ipython Notebooks, to be used in this workshop.

I have organized and mentored a couple of workshops at my University. This is a presentation from another workshop on Github, that I organized for Software Freedom Day'14: http://slides.com/rishabhshukla/deck--2.

Rishabh (~rishy)

Login to add a new comment.