Tuesday, January 28, 2020

AI Tuesday's 2020-01-28

Hi guys! I’m so thankful for the interesting information posted on the internet, so I’ve decided that I’m going to start something I call “AI Tuesday’s” and “CS Thursday’s”.

Every Tuesday, I will post a few of the interesting AI/machine learning related links I’ve ran into over the previous week with short commentary about why I find each link interesting and every Thursday, I’ll do the same, but the focus will be Computer Science & programming rather than AI. 

I’d love to get your thoughts on this, please comment below. Also if you have any links you’d like to see get a wider distribution, feel free to ping me directly.

Computational notebooks
Computational notebooks (Jupiter, Azure Notebooks, etc) are awesome, but they still have a lot of room for improvement. My biggest issue is getting tons of large data into them. Here is a preprint paper that dives into some of the problems and offers opportunities that may be explored to improve them.

Teach Machines Cause & Effect
More of a philosophical piece, but it’s true that much of our intelligence is in understanding cause and effect.

OpenAI Gym & 3D printed robot
Very well documented and really fun project about training a 3D printed robot to walk.

Open source speech separation & enhancement
Movies have always shown us the “enhance” functionality in photos and video. Now we have it in audio as well.

Pointscreen deepfake demo
Don’t believe everything you see. The technology is not perfect, but getting closer every day.

Banning facial recognition misses the point
True. We need to ban governmental & commercial surveillance worldwide. But how exactly do you do that?

Real-time object detection with Phoenix & Python
Nice step by step walkthrough. What are you going to do with it?

Simplifying Semi-Supervised Learning
Anything that helps minimize the amount of time we need to spend on data wrangling gets my upvote.

Methods for interpreting & understanding deep neural networks
We need all the help we can get trying to understand (and improve) why deep neural networks do what they do.

Genetic algorithms in production
I’ve wanted to use genetic algorithms in production since working on a tool to decide how to lay carpet around a house 20 years ago. Great use case.

#ai #ml #artificialintelligence #machinelearning #deeplearning #python #openai #deepfake





No comments:

Post a Comment