This course introduces us to the fundamentals of deep learning, taught by Prof BhikshaRamakrishnan.
The content starts off basic human brain design(neuron firing) and swiftly moves to MLPs, followed by CNNs, RNNs, sequence-to-sequence models, Attention, Autoencoders, VAE, GANs. Each topic is covered in great depth and details along with assignments aimed at implementation from scratch and on some real data. It had four total assignments focusing on MLP, CNN, RNN, and Attention. Each has two parts: In part 1 wewere expected to code everything from scratch in Numpy without the use of any automatic differentiation library like Pytorch. The intention is to understand everything from ground up.
In part 2: wewere expected to use Pytorch and compete on Kaggle by pushing the models in terms of performance. This covers the more practical aspect of making models work.
There are 14 timed quizzes in total which need to be completed over the weekend and a Final Project.
Our final project was aimed at Autonomous vehicle which is dependent on the perception system for understanding the surrounding while taking decisions and actions in the environment. But deploying these algorithm poses high-risk challenges in detection. With the availability of deep learning models, accuracy of 2D detection has been on rise, but not so for 3D.Many AV companies propose using LiDAR sensors which provide rich Pointclouds of ego vehicle surrounding environment. WE explored and proposed solutions based on centerness for 3D bounding box detection for LiDAR Pointclouds as inputs.
A short demo of the same could be found here:
Link to poster has been added to the report section.
The course website can be found here.