Practical Deep Learning with PyTorch

I have launched a course on deep learning with PyTorch that gets you starting with the PyTorch framework as well as understanding the transition from PyTorch fundamentals all the way to more complicated deep learning models. Sign up here:! The course is specifically designed such that if you...

Deep Learning for Self-Driving Cars and Medical Diagnostics by NVIDIA

I’ll be hosting a talk by Marek Bardoński, Senior Deep Learning Research Engineer, NVIDIA Switzerland. Our official partner is sponsoring the location at The Hangar by NUS Enterprise, and it will be held on 21 June, 1pm to 3.30pm. We will be touching on cutting-edge applications of deep learning for...

Comic Series: Life's Journey is Like Gradient Descent

If we keep learning and tweaking from our previous experiences, we would eventually reach the destination we desire.

Comic Series: Plain Vanilla

This is just a plain vanilla post.

REWORK Deep Learning Summit Singapore

I’ll be presenting at REWORK Deep Learning Summit Singapore on 27-28 April 2017! Abstract With every deep learning algorithm comes a set of hyperparameters. Optimizing them is crucial in achieving faster convergence and lower error rates. For many years, the majority of people in the deep learning...

Gloqo: Search for code for research papers on arXiv

This is basically “the Google” for searching code for computer science papers on arXiv. It’s currently concentrated on deep learning and machine learning. I started this companion website to to encourage reproducible research. You can easily search for code that are implemented by the authors themselves or often by...

Comic Series: The Future of Labelled Data

Weirdly, there is something similar offered by Scale

Comic Series: Support Vector Machines

Support Vector Machines are often called Largest Margin Classifier.

Comic Series: Convolutional Neural Nets

As a disclaimer, Derpina intentionally set the output to two classes: “pretty girl” or “not so pretty girl”. We do not know if she intentionally manipulated the labels that led to her findings.

Comic Series: Markov Decision Processes with High Rewards

Imagine yourself in a grid-like world with two end goals (absorbing states), what would you do? Would you reach for the drumstick or drop into the hole?

RSS Subscribe

See more posts from the News Archive.