Viewing profile — sherjilozair

Ace is actually uniquely designed to support that. Our training staff simply record their screens and mouse+keyboard events. We transform that into behavior cloning data to train t…

I'm the founder and CEO of General Agents. Happy to answer questions!

I’ve been trying to find a maintained typed scripting language. Thank you for building!

That's not the safety team.

So this is why my previous company had taken us to a retreat where we tried to brainstorm how to build AGI using post-it notes. We were given 5 minutes and then someone did a crude…

Good video content is created by first writing the script.

Yes, I can read it perfectly fine. It relies on a lot of notation that AI researchers are familiar with.

GIL can remain default on. Users can simply disable it for chosen parts of their program.

You know it’s trivial to add a global lock to any concurrent program, right? What would you lose if other people started writing performant code?

You don't have to use PICO-8's text editor. You can just open up the p8 file in any text editor and hack away. VS Code has an extension for .p8 files as well which does syntax high…

Look up iterator, __next__, __iter__.

DeepMind has already released the open source code and model parameters. The database makes it easier to access the predictions.

How would I include the definitions without the implementations?

It's Lua. They use Love2D.

FNA, Monogame, Heaps, Kha, and many other open source engines have closed source forks supporting consoles.

World Models is a good paper, but is definitely not the top three most significant RL papers.

But is autodiff combined with a blackbox jit a real solution? The jit either works for your new model or it does not. If it does not, you can do pretty much nothing about it, other…

CPC trains neural nets.

You only need a few seconds to read the title, and you can cut away 150 out of those 200 with that. If it's a paper simply not in your area of research, you don't need to bother. Y…

So in the case of AI/ML, the fashionable new thing is unsupervised learning, self-supervised learning, reinforcement learning, GANS, etc. The played out problems are image classifi…