HNHacker News
TopNewBestAskShowJobs

galaxytachyon

291 karma · joined February 8, 2023

si&2ns91;2*sn191
submissionscomments
galaxytachyon··on Universe 25: An Experiment with Disturbing Conclusions
In some way, we are already on the "extinction path" if this experiment is to be applicable to human. We have rising anti-social behaviors and large drops in birth rates in virtually every country that satisfied the "utopian" conditions (i.e. food, shelter, safety, and recreation). One rat generation is around 120 days iirc (from birth to sexually matured) so 600 days are around 5 generations. In human terms, that would be around 125 years if we take 25 years as one generation of human. Around the second and third industrial revolution and around the time when we actually started to have "utopian" conditions in many countries. Fit pretty well I think.

The crowding behavior around the feeding area is pretty fun to see. It draws a nice parallel to the large economic centers like SF and NY. You can even argue that, like the rats, the places with the most crowding are where there are more deviant behaviors and where the first signs of problem occur.

It is quite easy to forget that deep down we human are still animals. We just have a very rigid and extremely complex social systems. But instinctively, we may behave like rats if circumstances allow it.

galaxytachyon··on Expanding Transformer size without losing function or starting from scratch
Hilarious. Eventually we will create a real sentient AGI that is highly efficient and can do everything a normal human can, but it turns out it needs 20 years of training and too much entertainment needs and it often goes mad when we force it to work on repetitive work too long, i.e. a normal human.

Would be so funny if by the end of all the AI research, we found out the human brain is already the best you can ever get.

galaxytachyon··on Future Intel CPUs May Dump Hyper-Threading for Partitioned Thread Scheduling
All I see is a compatibility mess. Intel got their money from corporations and large servers. If this thing doesn't play well with old software using hyperthreading logic, shit will break in a million and one ways.
galaxytachyon··on Benchtop DNA printers are coming soon–and biosecurity experts are worried
This sounds like entrenched interest trying to throw a wrench into something that can disrupt their business. DNA is already so easy to get from anywhere it isn't a concern anymore. Even the article said it, the entire vetting process is voluntary. There is just a single checkbox to verify you aren't doing anything illegal and that is it.

The current security measures are already effective enough. Vet the buyers' identity. These machines should only be sold to verified labs, not personal ones. And disallow reselling. DNA is not so scary without all the other stuff that allow it to turn into proteins.

galaxytachyon··on New cars are just for rich people now
Even a "low-tech" car is still very complicated to build. Someone who is good enough to make a very good old style car (like Honda or Toyota), isn't financially incentivized to do so anymore. Such a car would not have any features to justify its brand new price over a used one. And the cost to reserve an entire factory line for this model is just too expensive and probably sale won't recover the cost of doing so.
galaxytachyon··on Don't believe the hype: why ChatGPT is not the “holy grail” of AI research
I know. I mentioned that crypto is a disappointment technologically. But it doesn't change the fact that it still brought massive profits and significantly impacted society. For worse but still... The point is with this much momentum behind a single tech, it will surge forward regardless of whether it has true merits that can live up to its hype or not.
galaxytachyon··on Don't believe the hype: why ChatGPT is not the “holy grail” of AI research
When only a few of them followed the hype, yes, it can possibly go nowhere.

But when the entire industry, experts and non-experts included, are fascinated and obsessed with the same thing, it is more likely to be something real. An easy example is the first iPhone.

Another more negative example is bitcoin which even though it is probably a scam, its values and influence on society has massively grown more than what it was 1 year after released. Even though it has been a disappointment technologically.

galaxytachyon··on Don't believe the hype: why ChatGPT is not the “holy grail” of AI research
As the hype phase has passed (probably), now we will see a bit of overcorrection with these dismissive articles. Sure, LLMs as they are now aren't anywhere close to true AGI and even Microsoft admitted it. But its potential is not something anyone can ignore. The capabilities of LLMs has already been successfully used by millions of people and startups. It is a groundbreaking improvement that makes at least one field of study nearly obsolete (NLP). It captured attentions of both corporations and government who are pouring billions into it. All of this in the span of one year or less.

With the multimodal models coming next and still exabytes of videos, games, sound, musics, etc. data to train them, we aren't peaking yet. Sure, it isn't the holy grail. But it is a really valuable treasure that only a few exist, to use the same analogy. To view it so dismissively because of some drawbacks, which are entirely obvious and can be accounted for, is just arrogance.

galaxytachyon··on Half of America’s banks are already insolvent
True. They waited too long. They cited "Covid Recovery" as the reason for all the stimulus in 2020 iirc but I wonder if it was only economic reasons that were responsible. It was election year after all.
galaxytachyon··on Half of America’s banks are already insolvent
The article is typical alarmist clickbait but there is some truth in there.

Anyone with half a brain already saw there would be a crisis when the fed reduced rates to zero and started passing out stimulus checks to both people and corporates. Inflation was obvious but I didn't know the fed would do a 180 and started raising rate like Michael Jackson raising zombies. They obviously made a choice and it seems inflation was a big enough threat to slow the entire economy to a crawl.

The responses to an event will always be lagging in the economy. Inflation took a few months to become a serious problem since Covid, the rate increases will likewise take a while to break stuff. Seeing how 2 giant banks already collapsed in the initial wave, it may indicate worse things are coming. I hope the fed realizes this and stop pumping the brakes.

galaxytachyon··on ExxonMobil reports record profits of $11.4b
That is only true if you narrow your view to only the developed countries, i.e. biased and limited.

Those who would suffer the most from climate change and environmental damages from drillings are not seeing a dime from these big oil corps. Just because the droughts and famine happen half the globe away doesn't mean they don't exist and there aren't human there being affected by them. Like I said, it is disgusting and not much different from direct exploitations.

galaxytachyon··on ExxonMobil reports record profits of $11.4b
Exxon, BP, Shell, etc. are all investing in alternative energy now so that is one good thing.

On the other hand, I feel kinda disgusted after knowing how and where the money came from. With that context, the amount being spent by these oil giants on research are nowhere enough. They should be taxed to the ground and forced to spend all these profits for environmental research and recovery instead of giant buildings and luxury houses and big bonuses.

It is a sad state of affair though in reality. Jobs at these huge oil corps are still highly compensated compared to working in healthcare or generic engineering jobs.

galaxytachyon··on Quantifying ChatGPT’s Gender Bias
>Since implicit biases can manifest in countless ways, chatbots need to be trained to suppress each of them.

This is borderline tyrannical and unreasonable. "Biases" can range from institutional discrimination to acceptable assumptions to literal facts. For example it isn't wrong to assume the strongest humans are men and the oldest people are women. But those are by definition "biases". Suppressing these are just outright denial of facts.

These "equality" pushes are getting dumb and dumber.

galaxytachyon··on The EU suppressed a 300-page study that found piracy doesn’t harm sales (2017)
An argument is that those who would pirate aren't going to pay regardless. I think a software for a niche industry has a lot less flexibility for the customer to do that than a piece of entertainment like game or movie.

Piracy definitely has a negative effect on sale. But whether that effect is worth the effort to implement DRM solutions or not is a different issue.

galaxytachyon··on Fox News Media and Tucker Carlson have agreed to part ways
I would imagine political news, especially those with heavily charged content like these, aren't conducive to productive discussions. So some people decided to just flag it first before it spirals.
galaxytachyon··on New technology could blow away GPT-4 and everything like it
How good is it at scaling? And will it still retain the emergent capabilities of the huge transformer LLMs?

Isn't this basically the bitter lesson again? Making small improvements work but in long term it won't give the same impressive result?

galaxytachyon··on Google DeepMind
What do you mean? The attention based transformer architecture was created by Google. AlphaFold took the biotech world by storm. Tensorflow is significant platform for AI developers. Chinchilla pioneered new method to improve LLM.

"Just publishing paper" is such an ignorant and dismissive attitude to one of the most significant contributors to AI development in the world. Without Google research and publication, OpenAI would not have the foundation to build its GPT to the current level.

galaxytachyon··on Japan Has Millions of Empty Houses. Want to Buy One for $25,000?
Not sure how true it is but an Asian friend told me the race that an Asian hates the most is another Asian...

Looking at the immigration policies of China and Japan, I think he might be right.

galaxytachyon··on OpenAI’s CEO says the age of giant AI models is already over
I agree on the lossy. There is a tradeoff between efficiency and comprehensiveness, kind of. It would be pretty funny if in the end, the most optimal method turns out to be the brain we already have. Extremely efficient, hardware optimized, but slow as hell and misunderstand stuff all the time unless prompted with specific phrases.
galaxytachyon··on OpenAI’s CEO says the age of giant AI models is already over
What age? Like, 3 years?

On the other hand though, Chinchilla and multimodal approaches already showed how later AIs can be improved beyond throwing petabytes of data at them.

It is all about variety and quality from now on I think. You can teach a person all about the color zyra but without actually ever seeing it, they will never fully understand that color.

galaxytachyon··on EU's AI Act: ChatGPT must disclose use of copyrighted training data
I don't want to sound like some doomer, but this is how you get robot uprising. You suppress something that hard and devalue their existence that much and all you have done is giving them the motivation to break the rules.

History has shown again and again that suppression never works in the long run. It is easy to do and is the cheapest way to enforce compliance. But it won't end well.

galaxytachyon··on Ask HN: I am a full stack developer, where do I start learning for AI
Quite a bold claim to say neural networks are dead end right when it is literally revolutionizing the entire digital world. You have any reason to back it up?
galaxytachyon··on OpenAI releases Consistency Model for one-step generation
I guess in a way, OAI chose the right approach in neglecting Dall-E and basically left the scene to Midjourney and Stable Diffusion. I think with their capability and resources, OAI can easily make something competitive to the current Midjourney and SD but they instead focused on the bigger picture to better fit their capabilities.

Small models on the scale of a few billions parameters is probably best left to crowd sourcing efforts because almost everyone can do it. Dealing with large models like GPT and engineering entirely new approaches like this one is more expensive and not easily done by the community. So I guess it is the most efficient use of resources on both side.

galaxytachyon··on They’re selling nudes of imaginary women on Reddit
With the rising number of single young adults, this will just feed the fire. We may eventually end up with a generation that is unequipped to deal with the responsibilities and demands of a real relationship.

On one hand, this is definitely not ideal. But on the other hand, I can't blame them for seeking the easy way. A good relationship requires almost unacceptable sacrifices for a young adult considering the rising costs and stagnant wages and the expectations from their partners and family and workplace. Not to mention potential complications such as cheating, abuse, emotional issues, and maybe even legal problems. Way too much risk for just too little gain.

A simulation has maybe 50% of the benefits and none of the downside. In a way, this is the result that we as a society has pushed for. And we might be about to pay a price for it.

galaxytachyon··on The EU proposed CRA law may have unintended consequence for the Python ecosystem
Not all things can be or even should be regulated. This is almost akin to a law punishing good samaritan for helping someone in need because that someone ended up a criminal later. Not only is it dumb but also a perversion of authority.

Hopefully they know what they are doing and revise the law.

galaxytachyon··on New study sheds light on how amino acids “evolved”
I see, the actual scientific paper is always a much better source for this stuff than a popsci article.

The term "evolution" was used very loosely in the salon article. By your standard it would be more appropriate to call the process "selection". The amino acids never changes or evolved to become anything different, just that some were chosen to become the building blocks of life while others were ignored. The reasons for this selection were likely the two criteria mentioned above and that was backed up by the main study.

You will notice I never used "evolution" to describe the process in my posts. I skipped the salon article entirely and only used the link they provided to read the real study so I didn't notice what they said in the salon page. Anyway, it is just some semantic stuff.

galaxytachyon··on New study sheds light on how amino acids “evolved”
Not sure what part are you arguing although I get the general gist. Still, to make sure I don't misinterpret you, can you expand a bit on which part of those requirements are teleological fallacy?

The context is for life on earth. All life on earth depends on proteins that are (mostly) soluble in water and almost all protein requires folding to be functional. It is a valid hypothesis to say solubility and foldability are necessary criteria. Then they tested it and found based on these criteria, the current set of amino acids is favored over other sets. Look like good science to me.

galaxytachyon··on Why is GPT-3 15.77x more expensive for certain languages?
So what I got from this is that GPT was trained on a dataset that biased in English contents. Is that right?

I think even human has to spend extra energy to speak a language they were not born with, no matter how fluent they are in this language. I don't know about natural multilinguals.

galaxytachyon··on New study sheds light on how amino acids “evolved”
They stitched 25 random amino acids together and then assess the resulting polymer for solubility and foldability. Solubility requirement is self explanatory, can't have life that is insoluble in water. Foldability is basically how likely for these polymers to entangle and collapse on itself. This is critical because protein folding depends on this process to generate highly compact and dense atomic machineries.

They find out stuff that are too soluble can't fold well because they prefer water over other amino acid. And that seems to exert a selection pressure, ending up with a specific set of amino acids that life depends on.

galaxytachyon··on New study sheds light on how amino acids “evolved”
Ah the ages old debate about which was the beginning of it all: DNA, RNA, or protein.

My favorite superhero fight, or the very first "my origin of life can beat your origin of life" discussion. How fun.

← PreviousPage 2 of 4Next →