Variational Auto-encoders Previously, my blog post spoke about generative adversarial networks, and how they work to create new samples of data from a learned distribution. In this post, we will look at another class of generative models based on auto-encoders,

Continue readingVariational Autoencoders (VAEs)

I am currently taking a course on visual learning and representation taught by Prof. Deepak Pathak. As a part of one of our assignments, we are implementing GANs from scratch. I have always been fascinated by the sort of results

Continue readingA Gentle Introduction to Generative Adversarial Nets (GANs)

Imagine if Robots could adapt to any environment they’re put into without explicitly being told how to do so. Similar to how toddlers learn how to walk by falling over again and again, what if we could impart this ability

Continue readingIntroduction to Reinforcement Learning

This is a text-based summary from the IROS 2020 Perception, Learning, and Control for Autonomous Agile Vehicles Workshop. The following post aims to summarize the key takeaways as well as explain some technical terms/concepts not elucidated in the tutorials. Some

Continue readingModular Deep-Learning Based Navigation Stack For Autonomous Driving

Abstract They present a robotic grasping system that uses a single external monocular RGB camera’s input. The object-to-robot pose is computed upon merging the output of two neural networks: one for calculating the object-to-camera pose, and another for calculating the

Continue readingIndirect Object-To-Robot Pose Estimation From An External Monocular RGB Camera