howrar

joined 2 years ago
MODERATOR OF
[–] howrar 1 points 1 day ago (1 children)

I can choose for myself. You can choose for yourself. We cannot choose for other people.

[–] howrar 1 points 1 day ago (3 children)

Yeah, they're two different things. The commonality is one person's actions negatively affecting other people. As the party that's being negatively affected, it makes no sense to say that it's not your problem just because you're not the cause of the problem. Being negatively affected by it makes it your problem.

[–] howrar 1 points 1 day ago (5 children)

If someone picks up a gun and accidentally shoots you in the foot, what are you going to say? "I'm not an idiot. I know gun safety. If you shoot me in the foot, that's your problem, not mine."

[–] howrar 1 points 1 day ago

Recovery time can vary a lot depending on the person, the particular muscle group how much volume you do, how hard you push, quality of your sleep, and a bunch of other factors. It's not wild to have arms that recover faster than average.

It's perfectly valid to have a fluctuating schedule too. It's not ideal, but life rarely cooperates to give us ideal conditions. I'd say that if changing it to a fixed schedule is too complicated or makes it less enjoyable and harder to adhere to, then don't do it. Based on what you've written, it seems like you do have a pretty well thought out plan on how to autoregulate and adapt to whatever your work schedule throws at you. That is in itself a rigorous plan. Not everything has to align with our seven day calendars.

[–] howrar 1 points 1 day ago

behoove

And just like that, Mike Isratel popped into my head to narrate for the rest of your post.

[–] howrar 1 points 1 day ago (7 children)

What makes you think it wouldn't? How do you inform yourself about the happenings of the world if not through the news? Or from people who read the news? And of those people, how often do you think they read past the headlines before jumping to a conclusion?

[–] howrar 3 points 1 day ago (2 children)

It has nothing to do with the meaning. If your training set consists of a bunch of strings consisting of A's and B's together and another subset consisting of C's and D's together (i.e. [AB]+ and [CD]+ in regex) and the LLM outputs "ABBABBBDA", then that's statistically unlikely because D's don't appear with A's and B's. I have no idea what the meaning of these sequences are, nor do I need to know to see that it's statistically unlikely.

In the context of language and LLMs, "statistically likely" roughly means that some human somewhere out there is more likely to have written this than the alternatives because that's where the training data comes from. The LLM doesn't need to understand the meaning. It just needs to be able to compute probabilities, and the probability of this excerpt should be low because the probability that a human would've written this is low.

[–] howrar 3 points 1 day ago (2 children)

I don't think we would've had so many lessons on this in school if it didn't need to be taught.

[–] howrar 8 points 1 day ago (2 children)

Have people just completely forgot how search engines work? If you search for two things and get shit results, it means those two things don't appear together.

[–] howrar 15 points 1 day ago (5 children)

A sentence saying she had her ovaries removed and that she is fertile don't statistically belong together, so you're not even getting that.

[–] howrar 2 points 6 days ago (9 children)

I'm talking about the problem with the article, not problems with society or the world or anything else. No one's stopping you from being upset at multiple problems at once. Unfortunately, I don't have the means of reaching the arsonist nor the author of the article to make my complaints, nor the means to experience anger (alexithymia), but I can communicate with the people of Lemmy and encourage people to actually think about what they read. It's also just a fun exercise to see how biased articles are written in the first place.

No one is making fun of the LGBQT community because of him.

Not making fun of. Promoting fear, and the idea that they are all dangerous. Rereading the comments, it's actually more an attack on anyone who supports the LGBTQ community than on LGBTQs. I'll quote some of them below for you.

A lipstick wearing arsonist. Sounds like your typical demokrat. (toadlick2)

Another Trans-Terrorist...that'll by the twit 40 Years in Jail. Good, throw away the key. (Pennsyltuckian)

This is what your typical Democrat looks like. (europa2832)

Look at this poster child of the liberal left.. these liberals are the most violent, the most bigoted and the greatest threat to our country.. they say they are for peace.. NO !!! They are not!!! Do u see Conservatives doing this?? Dont Give me that BS of January 6!!! (rockaway1)

This is the face of the left. And they are endorsing it. Nanny P and Schumer and all the crazies in that parties are endorsing violence. (Zee Chen)

I picked out the ones that are most explicit, but just about every comment is saying the same thing.

[–] howrar 13 points 6 days ago (3 children)

It would be more effective to "snitch" on doctors who deny such care.

5
Factorio Learning Environment (jackhopkins.github.io)
5
Open Sourcing π₀ (www.physicalintelligence.company)
 

https://bsky.app/profile/natolambert.bsky.social/post/3lh5jih226k2k

Anyone interested in learning about RLHF? This text isn't complete yet, but looks to be a pretty useful resource as is already.

 

Apparently we can register as a liberal to vote in the upcoming leadership race. What does it mean if I register? What do I gain (besides the aforementioned voting) and does it place any kind of restrictions on me (e.g. am I prevented from doing the same with a different party)?

 

An overview of RL published just a few days ago. 144 pages of goodies covering everything from basic RL theory to modern deep RL algorithms and various related niches.

This manuscript gives a big-picture, up-to-date overview of the field of (deep) reinforcement learning and sequential decision making, covering value-based RL, policy-gradient methods, model-based methods, and various other topics (including a very brief discussion of RL+LLMs).

 

If there's insufficient space around it, then it'll never spawn anything. This can be useful if you want to keep a specific spawner around for capture later but don't want too spend resources on killing the constant stream of biters.

10
submitted 5 months ago* (last edited 5 months ago) by howrar to c/[email protected]
 

I'm looking to get some smart light switches/dimmers (zigbee or matter if that's relevant), and one of the requirements for me is that if the switches aren't connected to the network, they would behave like regular dumb switches/dimmers. No one ever advertises anything except the "ideal" behaviour when it's connected with a hub and their proprietary app and everything, so I haven't been able to find any information on this.

So my question: is this the default behaviour for most switches? Are there any that don't do this? What should I look out for given this requirement?


Edit: Thanks for the responses. Considering that no one has experienced switches that didn't behave this way nor heard of any, I'm proceeding with the assumption that any switch should be fine. I got myself some TP Link Kasa KS220 dimmers and it works pretty well. Installation was tough due to its size. Took me about an hour of wrangling the wires so that it would fit in the box. Dimming also isn't as smooth as I'd like, but it works. I haven't had a chance to set it up with Home Assistant yet since the OS keeps breaking every time I run an update and I haven't had time to fix it after the last one. Hopefully it integrates smoothly when I do get to it.

 

This is a video about Jorn Trommelen's recent paper: https://pubmed.ncbi.nlm.nih.gov/38118410/

The gist of it is that they compared 25g protein meals vs 100g protein meals, and while you do use less of it for muscle protein synthesis at that quantity, it's a very minor difference. So the old adage still holds: Protein quantity is much more important than timing.

While we're at it, I'd also like to share an older but very comprehensive overview of protein intake by the same author: https://www.strongerbyscience.com/athlete-protein-intake/

 

Ten years ago, Dzmitry Bahdanau from Yoshua Bengio's group recognized a flaw in RNNs and the information bottleneck of a fixed length hidden state. They put out a paper introducing attention to rectify this issue. Not long after that, a group of researchers at Google found that you can just get rid of the RNN altogether and you still get great results with improved training performance, giving us the transformer architecture in their Attention Is All You Need paper. But transformers are expensive at inference time and scale poorly with increasing context length, unlike RNNs. Clearly, the solution is to just use RNNs. Two days ago, we got Were RNNs All We Needed?

 

Recordings for the RLC keynote talks have been released.

Keynote speakers:

  • David Silver
  • Doina Precup (Not recorded)
  • Peter Stone
  • Finale Doshi-Velez
  • Sergey Levine
  • Emma Brunskill
  • Andrew Barto
0
submitted 7 months ago* (last edited 7 months ago) by howrar to c/reinforcement_learning
 

OpenAI just put out a blog post about a new model trained via RL (I'm assuming this isn't the usual RLHF) to perform chain of thought reasoning before giving the user its answer. As usual, there's very little detail about how this is accomplished so it's hard for me to get excited about it, but the rest of you might find this interesting.

view more: next ›