📢Come to our NeurIPS session on Wednesday (4:30pm–6pm PT)! We show score-based generative models can be trained as a continuous normalizing flow (ScoreFlow) efficiently with a variational objective, achieving outstanding likelihood values that match best autoregressive models.

Attending #NeurIPS2022? Make sure to add our NeurIPS workshop on score-based methods ( https://t.co/gRoqSbmmgd) to your schedule! Our amazing panelists will be answering your questions—submit them now via this Google form: https://t.co/vOlzNvk58s.

Zoom in on the Andromeda Galaxy with Hubble. What looks like a smudge of light to the unaided eye is actually a vast galaxy containing an estimated 1 trillion stars. Credit: ESA/Hubble & NASA Full HD version on Youtube: https://t.co/uE6XQiyJ5e

An interesting collection of 800 free computer science video courses you can take online [access: https://t.co/qPPw11ZAl0]

Stare at the white dot for 10 seconds, then look at Van Gogh's Starry Night

The most dangerous lies we tell ourselves:

Everyone’s using ChatGPT. But almost everyone's STUCK in beginner mode. 10 techniques to get massively ahead with AI: (cut-and-paste these prompts👇)

7. NLP with Deep Learning by Stanford This course will enable you to build complex NLP solutions (in Pytorch) for a variety of NLP tasks. 👉

This one’s for you, Seattle. You stuck with us through thick and thin. We couldn’t be happier to celebrate this with the best fans in the world 💙

We’ve posted a new paper on how learning a new behavior leaves a “memory trace” of the learned behavior in neural population activity: https://t.co/3s40iN8V1J

"Learning alters neural activity to simultaneously support memory and action" Excited to share a new preprint, led by Darby Losey, with myself, Emily Oby, Aaron Batista, Byron Yu, Steve Chase, and team. 🧵 https://t.co/xMsBS5hOuP

Learning alters neural activity to simultaneously support memory and action https://t.co/IphPKBIw1s #biorxiv_neursci

I don’t love crypto VCs but Su and Kyle building up a multi-billion dollar fund from their closet is nothing short of remarkable. Su’s insight on podcasts have been some of the most valuable content I’ve found in the entire space. Hope they come out the other side of this. 🫡

TxnLab is pleased to announce that we are developing an #Algorand native naming service for .algo names. Our platform will enable users to send & receive $ALGO using a unique name that is easy to type and remember. 🧵👇

Excited to share we (@yulikeneuro @aaronbatista and others) have a perspectives paper on learning in neural population activity out now! "How learning unfolds in the brain: toward an optimization view" 📎 https://t.co/eRSEyg86TW (1/n)

The Art of Linear Algebra [pdf]

My current favt. projects on @avalancheavax 1) @yieldyak_ 2)@pangolindex 3)@novax_game 4)@AvawareAVE 5)@AvalaunchApp 6)@AVME_IO 7)@CanaryDex 8)@GajFinance 9)@GetKalao 10)@sherpa_cash 11) @BenqiFinance What about you? @FiN_Petch1 @darkseid89931 @Avalanche_In

Finished the 2nd Edition of "Data-Driven Science and Engineering: Machine Learning, Dynamical Systems, and Control" with Nathan Kutz today for Cambridge. Now w Python + MATLAB, extensive HW, and YouTube videos for all sections! (Code in Julia and R soon) https://t.co/m3S9fwy4y4

#China has billions of people. @Algorand is the future of finance. El Salvador is not the only country building on #Algorand. Algorand can be all things to all people groups in all nations. 🌎 🌍 🌏 $algo #AlgoAutumn #ElSalvador #BSN #blockchain #Finance

@KordingLab @BWJones @AdamHantman @jpillowtime @chrisXrodgers @gamaleldinfe You’re missing the enormous fact that rotations need to be coordinated across conditions to be “rotational dynamics”. Generating multiple trajectories like that will not give you coordinated rotations. No one claims rotations from a single trajectory.