Saved

16,933 bookmarks · page 285 of 339
2020-03-09 · topbots.com

Dissecting The Transformer

We saw how attention works and how it improved neural machine translation systems (see the previous blogpost), we are going to unveil the secrets behind the power of the most famous NLP models nowadays (a.k.a BERT and friends), the transformer. In this second part, we are going…

2020-03-09 · topbots.com

Decoding NLP Attention Mechanisms

Arguably more famous today than Michael Bay’s Transformers, the transformer architecture and transformer-based models have been breaking all kinds of state-of-the-art records. They are (rightfully) getting the attention of a big portion of the deep learning community and…

2020-03-09 · vitals.lifehacker.com

How to Quit Using a Barbell Pad for Squats

At most gyms, there’s a cylindrical pad on the floor somewhere near the squat rack. (If it’s not there, check the Smith machine.) When I first started squatting, I needed that pad on the bar—or thought I did. How else would I protect my bony neck? But the truth is, you don’t…

2020-03-09 · machinelearningmastery.com

Why Is Imbalanced Classification Difficult?

Imbalanced classification is primarily challenging as a predictive modeling task because of the severely skewed class distribution. This is the cause for poor performance with traditional machine learning models and evaluation metrics that assume a balanced class distribution…

2020-03-09 · distill.pub

Growing Neural Cellular Automata

Training an end-to-end differentiable, self-organising cellular automata model of morphogenesis, able to both grow and regenerate specific patterns.

2020-03-09 · vimeo.com

Murmuration

A chance encounter and shared moment with one of natures greatest and most fleeting phenomena. Facebook: Islands And Rivers For licensing information contact Dale Dobson - dale@iconiclinx.com

2020-03-09 · ryanholiday.net

Books To Base Your Life on (The Reading List) – RyanHoliday.net

“Can you recommend any books?” After being asked this question a thousand times, I put together a private email list of book recommendations. Each month, I send one email with 5 to 10 amazing books that I read, reviewed and think you’ll like. The email is simple and quick and…

2020-03-09 · towardsdatascience.com

Q-Learning

An early breakthrough in reinforcement learning — Off-policy Temporal-Difference control methods