The way it seems to work in Canada is that the government decides on a set of topics they want to fund that are fairly high level, and as long as your work falls in one of those categories, the grant gets approved. So the government doesn’t choose the specific drug to study. They choose which medical condition we want to try to treat, then they let the PIs tell them what they want to do and how it relates to those priorities.

They are icky [1].

References

[1] That LGBT propaganda law

So much tech support has moved to Discord. That’s worth keeping around.

Yes

You could say the same about people who used the early 2000s Google by entering full questions with natural language and clicking “I’m feeling lucky”. There are always going to be wrong ways to use a tool. But we’re discussing whether there exists a right way. And that right way includes verifying the information you receive, just like you would if you found it through a regular search engine.

The social and environmental costs are real. That’s not the criticism you gave and not what the responses are disagreeing with.

Willingness to look is a pretty important factor. LLMs reduce the personal cost incurred to look up information, similar to how search engines saved us from having to go to the library for every question we had.

We already had subreddit simulator for ages. This isn’t anything new.

The poll didn’t even ask a real question. “Yes AI or no AI?” No context.

No /s

This is a great variant to regular pizza when you happen to have the ingredients on hand. I do it all the time.

Might be a Boost bug, but the link doesn’t include anything past the hyphen.

Sorry, I misunderstood. I thought you were questioning the nutritional density of beans.

In any case, I haven’t done the literature review on this, but just based on literature I’ve encountered on the matter (ranging from scientific papers to pop-sci articles), everything that recommends a lower intake also specifies that they’re recommendations for the average sedentary person. If you have any sources to share that contradict this, please do share. I think I’m going to do a proper lit review on this soon, so it’ll be a good addition.

Or you can just look at the nutritional information for beans?

Yeah, there’s nothing we can do that literally harms no living creature. Any vegan who has given their choices a second of thought will acknowledge that it’s about harm minimization, not causing zero harm.

The main difficulty is in how many hyperparameters are involved in training an RL agent, high sensitivity of RL algorithms to those hyperparameters, and not having a good understanding of how to select them based on the properties of your task. This problem is exacerbated by the high sample complexity of RL. If something doesn’t work out, you don’t know if it’s because you chose the wrong set of hyperparameters or if you just haven’t trained for long enough.

I don’t know much about game design, but I do know that it’s a much more mature field than RL, so surely they have better tools than guessing and praying.

It is expensive, but it does work. We’ve already seen things work to a limited extent on StarCraft 2, Dota, and Gran Turismo, and those are all multiplayer games. The article seems to be talking about single player games, which simplified things a lot.

Game playing is not LLM. They’re game-specific reinforcement learning models. It’s not easy, but definitely doable with existing tech. Sony’s GT Sophy is a good demonstration on what they’re capable of.

I don’t know if you can describe it as “can’t be arsed” when their proposed solution is so much harder to implement.

There’s high demand for both RAM and GPUs coming from datacenters. Us regular consumers are just a tiny blip on their radar.

A little bit of rubbing alcohol and it comes right off