Mary Jo Feeney

8 Followers
63 Following
82 Posts
Seeking knowledge, insight, humor, and perspective. Business problem-solver and digital product leader. #Healthcare #Democracy #Leadership

Twitter is falling apart. Reddit is falling apart. Facebook fell apart ages ago. Meta is a trashfire. Instagram is baloney. Google can't even search for anything you want anymore.

You know what website still miraculously works?

Wikipedia.

You should donate to keep it that way.

As predicted, ML models suffer irreversible damage when you train them on generated data, a phenomenon these researchers are calling "model collapse". Uncurated datasets are effectively poisonous.

Who, apart from anyone who thought about it for a few seconds, could have predicted.

https://arxiv.org/abs/2305.17493v2

The Curse of Recursion: Training on Generated Data Makes Models Forget

Stable Diffusion revolutionised image creation from descriptive text. GPT-2, GPT-3(.5) and GPT-4 demonstrated astonishing performance across a variety of language tasks. ChatGPT introduced such language models to the general public. It is now clear that large language models (LLMs) are here to stay, and will bring about drastic change in the whole ecosystem of online text and images. In this paper we consider what the future might hold. What will happen to GPT-{n} once LLMs contribute much of the language found online? We find that use of model-generated content in training causes irreversible defects in the resulting models, where tails of the original content distribution disappear. We refer to this effect as Model Collapse and show that it can occur in Variational Autoencoders, Gaussian Mixture Models and LLMs. We build theoretical intuition behind the phenomenon and portray its ubiquity amongst all learned generative models. We demonstrate that it has to be taken seriously if we are to sustain the benefits of training from large-scale data scraped from the web. Indeed, the value of data collected about genuine human interactions with systems will be increasingly valuable in the presence of content generated by LLMs in data crawled from the Internet.

arXiv.org

Did she just google "army guy" and take the first picture she found?

For those who don't know this is a picture of Pat Tillman, outspoken critic of the Iraq War who called Bush a criminal.

He gave up an NFL career to enlist after 9/11 and bucked the major narratives at the time by being an atheist whose favorite author was Noam Chomsky.

He died from friendly fire in Afghanistan, widely believed to be fratricide due to his criticism of the war.

He'd have HATED Boebert.

Elmo's World

"Ya Done F*cked Up Now, Elmo"

TAGLINE: when the actual Inventor of "The Hashtag" has quit your social media platform"

#RejectedTVShowPilots
#HashtagGames
#irony #meta #birdsite #fediverse

(follow him here) 👇
@chrismessina

@DataDrivenMD my grandmother used to save peeps and then in July have an “Easter BBQ” and toast them over a campfire. That’s the best way to eat them—the sugar caramelizes and they’re delicious. Just make sure they cool first!
What if meetings are just a conspiracy by extroverts to suck the life force from regular people?
This song really resonates today. Mark Erelli, Rosanne Cash, and friends. https://fb.watch/jy01Yyjvdm/?mibextid=YCRy0i #schoolshooting #nashvilletn
Mark Erelli, with Rosanne Cash, Sheryl Crow, Lori McKenna, Anais Mitchell & Josh Ritter - "By Degrees" (Official Lyric Video) | 100% of the proceeds from downloads & streams of this song will go to Giffords: Courage To Fight Gun Violence. | By Mark Erelli | Facebook

6.503 views, 73 likes, 40 loves, 9 comments, 104 shares, Facebook Watch Videos from Mark Erelli: 100% of the proceeds from downloads & streams of this song will go to Giffords: Courage To Fight Gun...

Actual footage of my last design thinking session… 🫢
I got Queen Bee!

How many words can you make with 7 letters?

In 1968, at 30 years old, Lynn Conway transitioned. In doing so she lost her wife, her children, and her job at IBM. She continued on, living authentically as her true self and continued her career as an electral engineer. In 1978 she became an associate professor at MIT and taught a course in VLSI (very large scale integration) that became the basis of the Mead-Conway VLSI Design Methodology, changing how we design integrated circuits. In 1985 she became a professor of electrical engineering at the University of Michigan, and later the associate dean of engineering.

32 years after transitioning, Conway came out publicly as a transgender woman. Since than she has been a advocate for transgender people in the tech sector. In 2020 IBM formally apologized for firing Conway for being trans, over 50 years after the fact. Just a bit of #transHistory I learned today.

#history