Thank you @samjoiner , @love_persis , @inari_ta and @sam_learner and @johnthornhillft for this compelling multimedia story about how artificial intelligence methods could help unlock incredible scientific discoveries about the rich and complex lives of our co-species on Earth.
The same technology that is powering ChatGPT could allow us to reach into the non-human world, and begin to understand — even speak — animal languages, or so researchers hope. Here’s how: on.ft.com/4aYsmU5
Could exponential advances in AI help us break down the interspecies communication barrier? Check out our newly published technical roadmap to see how we’re approaching the challenges and opportunities on the road to decode. earthspecies.org/blog/esp-te…
Whispering orcas, turtle hatchlings communicating through their shells, coral larvae using sound to navigate to their home reefs… Could learning to listen to other species help solve for global challenges? Check out our latest blog and watch the video! earthspecies.org/blog/using-…
Exciting opportunity for an AI Research Scientist to join our growing team and help apply machine learning techniques to the challenge of understanding the language of other species! earthspecies.org/jobs/jobs-s…
Happy Mother’s Day! After a YEAR of painting, my animated short Chamoe is now live. See the full film at:
vimeo.com/colbay/chamoe
I celebrate all mothers with this little film! I hope it brings 2 min of light and warmth to many moms out there today ❤️ #chamoefilm#MothersDay
Our work on end-to-end source separation for animals—the ability to isolate the individual calls when multiple bats, dolphins, and macaque monkeys are vocalizing—is out in Scientific Reports and Github
Blog: earthspecies.medium.com/towa…
Paper: nature.com/articles/s41598-0…
🧵⤵️ 1/n
There is a lot I ❤️ about @fastdotai v2, but probably my favorite thing is the flexibility it allows in reading in data.
The example in chpt #6 of the fastbook github.com/fastai/fastbook/b… where you start with an empty DataBlock is 🤯.
If you would like to see a RL example of this,
Using the out of the box functionality from
@fastdotai v2 to troubleshoot a model that wouldn't train...
... and fixing the situation by implementing stratified sampling of epochs of arbitrary length
So little code needed! 😺
github.com/earthspecies/sper…
With recorded COVID-19 cases (outside china) so eerily matching an exponential, I couldn't resist making a primer on exponential/logistic growth. At least 3 counterintuitive things about this kind of growth seem worth putting into the discussion.
piped.video/Kas0tIxDvrg
As data scientists, we (with @math_rachel) have done our best to look at the data around covid-19, and what it means to you and your community.
Our view: it is appropriate to be very concerned, and significantly change your lifestyle, right now.
1/13
fast.ai/2020/03/09/coronavir…
Just love today's season opener of @NPRinvisibilia - a moving backstory behind @earthspecies, the open-source nonprofit on a mission to decode animal communication. So proud of @aza and @bs, and blown away by how beautifully @aspiegelnpr told this story.
Hey Everybody. So today our season starts with a story that - when I was reporting it - hit me really hard and changed a bunch of my behaviors day to day. I’m really proud of it and hope that it will make some small amount of difference. npr.org/programs/invisibilia…
Last words of every inmate executed in Texas since 1984 – Browsing public language corpora and ran across this one. Profundity coupled with rote tabular data.
tdcj.state.tx.us/death_row/d…
I’m quoted in this. I think software used to make decisions such as hiring, firing, parole, prison sentences, etc. should be required to be open source. politico.com/agenda/story/20…