Chitti: Reinforcement Learning for Virtual Home Tasks
Description
Have you ever wondered how Virtual Assistants like ‘Siri’ and ‘Alexa’ work? Or Chitti from ‘Robot’ was programmed to perform complex tasks? In this ambitious project, we aspire to build a virtual home tasker using Reinforcement Learning techniques. We’ll first be training agents (virtual robots) in a highly efficient photorealistic 3D simulation before transferring the learned skills to reality. This will be a transition from ‘internet AI’ based on static datasets to AI within realistic environments. This will lead to a doorway to developing an android humanoid robot, which can help people in their household work and generalize it to outdoor and indoor tasks. The video below describes the environment and the basic actions that the machine will perform and would be trained using Reinforcement Learning techniques.
Application Process
To apply for this project one must attempt to solve a (simple) selection task and state why you want to be part of the project. The task involves parts:
- Implementation of a simple Neural Network from scratch (using Pytorch),
- Additional questions which are related to the project itself. There is no right or wrong answer to these questions and its basically to understand your thought process.
Interest Form
Asp. PM WhatsApp Group Link
Application Link
TIP: Deep Learning was invented to mimic how a neuron/human learns (atleast to a certain degree). Keep this in mind while answering the questions above. Your answers need not be technically accurate (but it would be appreciated) and can be loosely worded, but they MUST have some intuition backing them.
Submission Deadline
Completed applications are to be submitted by 14th May 2022 11:59PM!