anna-louise-ellis.bsky.social
Discovery for Science.
42 posts
37 followers
69 following
Regular Contributor
Active Commenter
comment in response to
post
Also Microsoft
comment in response to
post
THEY won’t be writing it….
comment in response to
post
🤣👏
comment in response to
post
Medical science …
comment in response to
post
Also, with respect to all the machine learning papers stolen, generally credit is given to the impact of your machine learning research as it is used, which in reality means getting paid a living wage because your research is useful.
comment in response to
post
There are paywalled journal subscriptions from many scientists in the corpus that have been pirated.
comment in response to
post
Further it’s taken the Machine Learning papers we ML scientists write and has used them to advance its architecture. Meta have also refused to sign up to the AI Security Institute meaning, unlike the 60+ countries that did (including UK - bar one part about sharing resource), there are 0 guardrails
comment in response to
post
Hard for us Machine Learning Scientists to see our papers sucked into, and no doubt assisting development of these models 😡
comment in response to
post
Yep - us scientists too 😭
comment in response to
post
Those of us scientists who are now not going to get referenced, yet the work be used generally, are funding the billionaires and getting poorer all the while.
It’s not about libgen. It’s about Meta’s use of libgen.
comment in response to
post
‘Causal’ component architectures :)
comment in response to
post
US is missing a real trick here.
Responsible AI techniques lean in to the development of ‘Discovery AI’, which surely is about the ‘flourishment of humans’.
Happy to have a conversation around Causal AI being the next frontier for ML and the techniques being developed
comment in response to
post
Yeah, they didn’t sign up to anything at the AI Action summit mid February. Think we are missing just one around sharing resource but the rest we’ve agree with.
comment in response to
post
Make more ‘discovery-athons’ as opposed to hackathons
comment in response to
post
EBay
Argos (uk)
Etsy
comment in response to
post
Specifically working on explainable AI techniques for weather and climate, over the pond here in the UK and very much appreciate the challenging work you are doing!
comment in response to
post
👏🥳👏
comment in response to
post
🤣. NVC++ anyone?
comment in response to
post
Pics or it didn’t happen …
comment in response to
post
And, for anyone using AI for science…this…
comment in response to
post
Amy - just wanted you to know that across the pond we hear you.
comment in response to
post
Read more about how this new framework addresses the challenges facing science diplomacy today, including increased economic competition; rapid technological advancements like artificial intelligence; and the rise of non-state actors including “tech titans.” royalsociety.org/science-diplomacy
comment in response to
post
So which chimes with your own values
comment in response to
post
Think Microsoft have actually grown a back bone and sticking firm on their DEI
comment in response to
post
I wish I stuck at it now 😭
Shall get back in the game
comment in response to
post
I transferred 3 mnths worth of mail (been using google for 25 yrs), to proton.me. There’s tools online to support this but I simply did a clear up on my gmail, then used proton.me’s service to port 3 months worth over (you can do longer time frames). Slowly changing my apps email contacts over.
comment in response to
post
Taken me about 6 hours (I’ve been google for 25 years). It WAS painful and I’m still getting used to proton, 4 days in, but oh the joy of it 😁
comment in response to
post
However google HAVE removed their DEI policies. Unlike Apple
comment in response to
post
ML for science NEEDS the AI Safety Institute.
Otherwise you are simply reducing the tooling to risk averse skills. Very limiting outlook by all accounts.
comment in response to
post
comment in response to
post
Mythical man month anyone?
comment in response to
post
Not just physicists!
comment in response to
post
Duck duck go and Proton so far for me but lots to unhook after 25 years of use😭
Feels like a divorce 🤣
comment in response to
post
Tell the world this is how we and weather fit together
comment in response to
post
Oh yes!!
comment in response to
post
😳😬🫂