Introduction to Deep Learning for Natural Language Processing
Submitted by Nischal HP (@nischalhp) on Friday, 29 April 2016
Section: Workshop Technical level: Intermediate
This workshop will provide an introduction to deep learning for natural language processing (NLP). It will cover some of the common deep learning architectures, describe advantages and concerns, and provide hands-on experience.
We would cover the following:
- What is deep learning?
- Motivation: Some use cases where it has produced state-of-art results
- Basic building blocks of Neural networks (Neuron, activation function, back propagation algorithm, gradient descent algorithm)
- Supervised learning (multi-layer perceptron, recurrent neural network)
- Introduction to word2vec
- Introduction to Recurrent Neural Networks
- Text classification using RNN
- Impact of GPUs (Some practical thoughts on hardware and software)
Broadly, there will be three hands-on modules
- A simple multi-layer perceptron - to understand basics of neural networks (everything will be coded from scratch)
- Vectorization hands-on using Word2vec
- A text classification problem and a text generation problem: This will be solved using Recurrent Neural Networks.
Laptop with python packages installed.
- Anaconda for Python 2
- Keras -
pip install keras
- Lots of enthusiasm
Nischal is co founder and Data Engineer at Unnati Data Labs who enables the Data Scientists to work at peace. He makes sure that they get the data they need and in the way they need it. Previously he has built, from scratch, various systems for E-commerce like catalog management, recommendation engines and other systems that amass a lot of data, during his tenure at Redmart.
At SAP Labs, Nischal has built various data crawlers, intention mining systems and laid down initial work on an end to end Text Mining/Analysis Pipeline. The majority of his work, however, was centered around building a system that gamified technical indicators in a product for the Fintech domain.
He has conducted workshops in the field of Deep learning across the world. He is a strong believer of open source and loves to architect big, fast and reliable systems. In his free time, he enjoys psychedelic trance and travelling to remote places.
Raghotham is a Data Scientist who can work across the complete stack. Previously, at Touchpoints Inc., He single handedly built a data analytics platform for a fitness wearable company. With Redmart, he worked on the CRM system and has built a sentinment analyzer for Redmart’s Social Media. Prior to Redmart and Touchpoints, Raghotham worked at SAP Labs where he was a core part of what is currently SAP’s framework for building web and mobile products. He was a part of multiple SAP wide events helping to spread the knowledge both internally and to customers.
Having found deep love for data science, neural networks and the passion for teaching, Raghotham has conducted workshops across the world. Apart from getting his hands dirty with data, he loves travelling, Pink Floyd and masala dosas.