Python for Artificial Intelligence

Yash Sherry (~yash10)


79

Votes

Description:

Python has been one of the easiest and most flexible language to use in the domain of Artificial Intelligence for years now. Since its inception , the community has witnessed a lot of emerging libraries and packages being released to make the working even easier. There is hardly any other language with such great support from a very healthy community of computer enthusiasts.

This talk is primarily showcasing the use of emerging python libraries in domains of various fields of artificial intelligence. The fields we would be covering would be Speech Recognition, Artificial Neural Networks, Deep Reinforcement Learning and finally Reinforcement Learning.

We would be covering these ideas with understanding codes for a project in each domain:

 1.  Voice Recognition

Libraries : Speech Recognition 3.4.6

Project : Voice controlled shell

Description : We would develop and understand a small voice controlled shell, that does most of the tasks that your normal linux shell does.

  2. Neural Networks

Libraries : Sklearn, nolearn, numpy

Project : Image Classification

Description : We will discuss and implement a deep neural network for image classification using popular datasets.

  3. Reinforcement Learning

Libraries: Theano

Project : Google DeepMind's DQN

Description : Discussion and understanding of the popular DQN code by Google Deepmind , and running the python implementations of the same.

Prerequisites:

Basic understanding of Neural Networks, use of Numpy and other related libraries. Setup : numpy,Speech Recognition,Scipy,Theano

Content URLs:

Here is a link to the videos of these projects running : https://www.youtube.com/channel/UCcedwpxpmWggI_Wcfx3ijxQ

All videos would be added by 1st July. I would be adding the link to the presentation and the codes too by 2nd July.

Speaker Info:

I am a second year student at IIIT-Delhi, majoring in the field of Computer Science. I have a decent amount of experience in research activities and I am a core member of two Korean research labs , Irisys and Optimede. Apart from this, I have worked as a research intern at Stanford in the domain of Crowd Research under Prof Michael Bernstein in the domain of Data Science.

I am currently working with Carnegie Mellon University in the field of Reinforcement Learning , wherein we have developed an algorithm faster than the current DQN code by Google's DeepMind. We are publishing our idea shortly in a prestigious conference.

Speaker Links:

Gmail : yash14123@iiitd.ac.in LinkedIn : https://www.linkedin.com/in/yash-sherry-63ab8aaa Project demos : https://www.youtube.com/channel/UCcedwpxpmWggI_Wcfx3ijxQ Open Source : https://github.com/theaverageguy/

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