physics-to-ml-pipeline:latest
England
Joined January 2009
- Tweets2.1K
- Following710
- Followers230
- Likes1.6K
Patrick retweeted
This is huge: Llama-v2 is open source, with a license that authorizes commercial use!
This is going to change the landscape of the LLM market.
Llama-v2 is available on Microsoft Azure and will be available on AWS, Hugging Face and other providers
Pretrained and fine-tuned models are available with 7B, 13B and 70B parameters.
Llama-2 website: ai.meta.com/llama/
Llama-2 paper: ai.meta.com/research/publica…
A number of personalities from industry and academia have endorsed our open source approach: about.fb.com/news/2023/07/ll…
Patrick retweeted
there is a huge difference between knowledge and expertise. individual humans are not so great with acquiring and retaining knowledge, but are remarkable at developing deep expertise. language models seem to be the exact opposite. they are amazing in the knowledge part.
Patrick retweeted
I think it's a horrible idea to ask for licenses to train models, this will reduce number of open-source models and will give big corporations an unfair competitive advantage to train closed-source models which will not be transparent at all and companies will have to sacrifice a lot of privacy since they don't own their models. Monopolies aren't good.
If a system affects people, people have the right to know more about it (biases and so on), given we know companies are using these models in all kinds of bad ways (they even use it for candidate screening 😔)
Patrick retweeted
There’s one existential risk I’m certain LLMs pose and that’s to the credibility of the field of FAccT / Ethical AI if we keep pushing the snake oil narrative about them.
Patrick retweeted
We release a open letter to the EU Parliament supported by Prof. Jürgen Schmidhuber,
Prof. Sepp Hochreiter,
Prof. Bernhard Schölkopf, the European Laboratory for Learning and Intelligent Systems, the German AI Association ( >400 businesses ) & many more:laion.ai/notes/letter-to-the…
Patrick retweeted
I'm seeing a lot of people confused about this - asking: what exactly is the problem here? That's a great question!
Let's use this as a learning opportunity and dig in. 🧵
Patrick retweeted
🧵 Some thoughts about the recent release of Galactica by @MetaAI (everything here is my personal opinion) 👀
Let's start with the positive / What went well
Patrick retweeted
All of twitter’s ML Ethics, transparency, and accountability team (except one). was laid off today. So much for that responsible ML initiative, huh?
Patrick retweeted
PEOPLE ASSUME CORRELATION EQUALS CAUSATION BUT LIKE ICE CREAM SALES DON’T CAUSE SHARK ATTACKS RIGHT THEY’RE CORRELATED BECAUSE THEY BOTH GO UP IN THE SUMMER WHEN IT’S HOT OUT BUT IT’S NOT MY JOB TO TEACH THE WORLD ABOUT A SPURIOUS RELAT—
Patrick retweeted
👀 Sneak a peek at the deepest & sharpest infrared image of the early universe ever taken — all in a day’s work for the Webb telescope. (Literally, capturing it took less than a day!) This is Webb’s first image released as we begin to #UnfoldTheUniverse: nasa.gov/webbfirstimages/
ALT The background of space is black. Thousands of galaxies appear all across the view. Their shapes and colors vary. Some are various shades of orange, others are white. Most stars appear blue, and are sometimes as large as more distant galaxies that appear next to them. A very bright star is just above and left of center. It has eight bright blue, long diffraction spikes. Between 4 o’clock and 6 o’clock in its spikes are several very bright galaxies. A group of three are in the middle, and two are closer to 4 o’clock. These galaxies are part of the galaxy cluster SMACS 0723, and they are warping the appearances of galaxies seen around them. Long orange arcs appear at left and right toward the center.
Patrick retweeted
Follow @OzKaterji. He's braver than I am and better looking. I've challenged Putin to his face. Between the two of us, you get a not bad picture of Putin's war.
Patrick retweeted
There is a lot of concern/confusion about vaccine effectiveness against the delta variant. How effective are the vaccines against Delta & how to interpret real-world observational data? So much misinformation is being circulated, so this thread brings key data together. 🧵(1/n)
Patrick retweeted
Things no-one else is crazy enough to be doing right now: final prep for a trip to answer one of the great unanswered questions: just how far can you travel by bus from London in 24hrs? I *think* I know on paper, but in practice…? Follow along from 3am as I slowly go mad. #bus24
Patrick retweeted
In 1967, Jocelyn Bell Burnell made a discovery that altered our perception of the universe. But the world wasn’t ready to accept that a breakthrough in astrophysics could come from a woman. nyti.ms/3iR7xQj
Patrick retweeted
NEW: probably the most important Covid chart I’ve made
As Delta goes global, it’s a tale of two pandemics, as the heavily-vaccinated Western world talks of reopening while deaths across Africa and Asia soar to record highs
My story with @davidpilling ft.com/content/fa4f248a-a476…
Patrick retweeted
SCOOP:
Public Health England has presented the first real-world data on vaccine efficacy against B.1.617.2, the variant first found in India.
Efficacy against symptomatic B.1.617.2 was 81% after two doses, much higher than many have feared.
Story: ft.com/content/a70d423a-7d7c…
