Back To Schedule
Saturday, October 20 • 1:30pm - 2:00pm
Deep Learning for NLP (Natural Language Processing)

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!

Feedback form is now closed.
As my main focus is on applying Deep Learning techniques in NLP (Natural Language Processing) field, I would love to speak about couple of my research projects which I did with New York University. First one was on Reading Comprehension task, where we tried to beat the state of the art models with experiments on attention based RNN (Recurrent Neural Network) models. We performed our task on two Daily Mail reading comprehension data set. Second one was with Prof. Kyunghyun Cho, where our main goal was to develop a system capable of summarizing content for user query. Here as well we tried to solve the problem with variety of convolution and recurrent neural networks with attention.

I would like to start my session by introducing the development of Deep Learning since last few years in NLP domain. I can control the technical level of my speech based on the level of audience.

More information on these research projects is available on these links below:
Summarization : https://github.com/up276/NLP2K16/blob/master/Two_Layer_Attention_Reading_Comprehension/NLP_Project_Paper_up276_vec241.pdf

Reading Comprehension: https://github.com/up276/QueryBasedSummarizationNLP/blob/master/IndependentStudyReport_vec241_up276.pdf

Other projects: https://cims.nyu.edu/~up276/info.html

avatar for Urjit Patel

Urjit Patel

Data Scientist, Praedicat Inc | New York University
I hold a Master’s degree in Data Science from New York University. I am an active researcher in Deep Learning and NLP domain. I was part of Prof. Kyunghyun Cho's (CIFAR Azrieli Global Scholar) NLP research group at NYU. I was a Teaching Assistant for Prof. Yann LeCun (Director of... Read More →

Saturday October 20, 2018 1:30pm - 2:00pm
Ballroom # 403B