chriskanan

11 ms·

91.
by chriskanan 3y ago
I read the book "Getting what you came for" before starting my PhD and took that to heart for graduating. It basically argued one does not need to do their best work during their PhD as that's the start of one's career.
92.
by chriskanan 3y ago
The article is about how ranchers and the meat industry is lobbying against lab grown meat, since it is a potential competitor and threat.
93.
by chriskanan 3y ago
This could potentially increase the number of useful questions being asked. My spouse is a physician. Recently she had to answer a phone call from a patient in the middle of the night where they were asking a question from a party about som
94.
by chriskanan 3y ago
I 100% agree. They also apply rules inconsistently, where the handful of times I've tried to make an article on well-known scientists (h-index 70+) years ago, the editors rejected it despite some secondary sources. On the other hand th
95.
by chriskanan 3y ago
Many people have an occasional year where they make very high incomes, for exam when selling a house or even fictional paper gains when exercising stock options as a startup employee. So, given that they aren't making huge salaries but
96.
by chriskanan 3y ago
Here are a couple papers related to this topic, but the primary focus has been on showing that LLM embeddings correlate with neural activity better than alternatives. But given one has neural recordings, one could train a model to take them
97.
by chriskanan 3y ago
The trades are much better protected in the next decade than many jobs requiring a college education. Robots are expensive, prone to breaking, and progress has been slow. AI is going to be disrupt many industries, just as tech has in the pa
98.
by chriskanan 3y ago
It depends on context. Often reviewers are chosen because they work specifically in that area, and of course they know their work best. I've often done this as a reviewer in appropriate situations, where I point out closely related wor
99.
by chriskanan 3y ago
Regarding inflating the number of authors, it is especially bad in medicine, where I've observed a lot of names being added to papers for "political" reasons, despite the "author" playing no role in the paper. Some
100.
by chriskanan 3y ago
I've observed this in multiple AI niches. In some cases I've emailed people saying they ignored very similar work and failed to cite it, and in at least some cases, they were apologetic and said they would update the arXiv version
101.
by chriskanan 3y ago
Other skin cancers have a low mortality and can mostly just be chopped out because they don't tend to invade other tissues and spread throughout the body. Most skin cancer deaths are from melanoma because it often invades and spreads.
102.
by chriskanan 3y ago
Sensory and motor systems have critical or sensitive periods early in life and then the neural network is pruned and the critical period closes. This is why second language acquisition is much easier at younger ages. There are some drugs th
103.
by chriskanan 3y ago
Regarding new competitors, I assume the upfront capital expenditure for many of the areas would be prohibitive for raising enough money from typical sources, especially since the existing industry could just lower costs once a new competito
104.
Galaxy Discovered with Seemingly No Stars Whatsoever(sciencealert.com)
4 points by chriskanan 3y ago|0 comments
105.
by chriskanan 3y ago
Delta doesn't have any currently, although they ordered a bunch of the MAX family that will be delivered from 2025 - 2029. It will probably be impossible to avoid them, long-term.
106.
by chriskanan 3y ago
This would totally derail open-source LLMs and establish a SV monopolies since they have sufficiently deep pockets to probably license an awful lot of material. Typically, the model is trained by "reading" each document only a sin
107.
by chriskanan 3y ago
What made me very fast was playing MUDs during high school (25+ years ago, and specifically Arctic MUD, http://mud.arctic.org/ ). Have to type fast or your character is going to die. Once you get used to it, an ergonomic ke
108.
by chriskanan 3y ago
Regarding why we don't see them in stores, in part I think that's related to them being perishable but also because gooseberries and currants were both banned in the USA until 1966 because it carried a disease that hurt the white
109.
by chriskanan 3y ago
Are you fertilizing? I have one in a pot that yields probably 200 berries per year. The issue is the thorns. I bought a couple thornless varieties last year. Hopefully will get some fruit this year from those.
110.
by chriskanan 3y ago
The issue is that this has caused severe shortages. The only new card with NVLink is at the very highest end and when trying to get a quote recently, I was told there was a 13 month delay in shipping. But if I don't need NVLink, just a
111.
by chriskanan 3y ago
Phi2 was created by Microsoft, that's probably why that was acceptable.
112.
by chriskanan 3y ago
This. It's insane to have to pay huge taxes when exercising stock options as a startup employee when many times the company goes bust. One should only pay taxes on real money received. There are many horror stories.
113.
by chriskanan 3y ago
The developers of Pytorch put out an article about this a few years ago and they admitted Julia would make sense from a pure ML perspective but the weight of the Python ecosystem was too critical. I wonder if GPT-4 could be used to efficien
114.
by chriskanan 3y ago
While I certainly agree that OpenAI isn't open and is effectively controlled by Microsoft, I'm not following the "groupthink" claims based on what just happened. If I'd been given the very fishy and vague reasons th
115.
by chriskanan 3y ago
I could really use some advice for making some deep learning hardware purchases for my university lab. It has been a pretty painful experience speaking to the companies due to shortages and because of all the limitations of non-H100 chips.
116.
by chriskanan 3y ago
Having previously been an AI scientist hiring manager, that's around the starting salary for an AI PhD fresh graduate in the USA, which does not include equity.
117.
by chriskanan 3y ago
The brain has a lot of skip connections and is massively recurrent. In a sense, the brain can be thought of as having infinite depth due to recurrent thalamno-cortical loops. They do mention thalamno-cortical loops in the paper, so I think
118.
by chriskanan 3y ago
This is what I've been calling it as well. Confabulation is much more appropriate as a term.
119.
by chriskanan 3y ago
The appropriate term from psychology is confabulation. Hallucinations are misinterpreting input data, but confabulations are plausible sounding fictions. https://en.m.wikipedia.org/wiki/Confabulation
120.
by chriskanan 3y ago
He mentions that there is a conflict of interest with recommending peer reviewers. While I agree this can be abused, I've often run into cargo cult science in AI when publishing something that is valid, novel, and in my opinion advance

More ›