Harini Sudha's Avatar

Harini Sudha

@yoursharini

PhD student at Berens lab @hertie-ai.bsky.social @unituebingen.bsky.social | #ELLIS & #IMPRS-IS PhD student | ML for science, NeuroAI | ex: @fz-juelich.de KTH IISERP

144
Followers
767
Following
2
Posts
19.11.2024
Joined
Posts Following

Latest posts by Harini Sudha @yoursharini

πŸŽ“ We're hiring two PhD students in #MachineLearning for my new group at the Hertie Institute for AI in Brain Health, University of TΓΌbingen!

Work on unsupervised learning, #DeepLearning, and #Neuroscience in a vibrant research environment.

πŸ‘‡ Details below

13.11.2025 09:41 πŸ‘ 8 πŸ” 3 πŸ’¬ 2 πŸ“Œ 1
Preview
The official home of the Python Programming Language

TLDR; The PSF has made the decision to put our community and our shared diversity, equity, and inclusion values ahead of seeking $1.5M in new revenue. Please read and share. pyfound.blogspot.com/2025/10/NSF-...
🧡

27.10.2025 14:47 πŸ‘ 6414 πŸ” 2756 πŸ’¬ 125 πŸ“Œ 452
Post image

New #NeuroAI preprint on #ContinualLearning!

Continual learning methods struggle in mostly unsupervised environments with sparse labels (e.g. parents telling their child the object is an 'apple').
We propose that in the cortex, predictive coding of high-level top-down modulations solves this! (1/6)

10.06.2025 13:17 πŸ‘ 22 πŸ” 6 πŸ’¬ 1 πŸ“Œ 2

Check out our new paper!
Vision models often struggle with learning both transformation-invariant and -equivariant representations at the same time.

@hafezghm.bsky.social shows that self-supervised prediction with proper inductive biases achieves both simultaneously. (1/4)
#MLSky #NeuroAI

14.05.2025 12:57 πŸ‘ 45 πŸ” 13 πŸ’¬ 2 πŸ“Œ 0

"Our results demonstrate that the better a model understands the data, the more effectively it can compress it, suggesting a deep connection between understanding and compression." - such a cool implication

03.05.2025 15:02 πŸ‘ 20 πŸ” 4 πŸ’¬ 3 πŸ“Œ 0

complete blog post is now live! more tips, and reframed into 2-step process: (1) get accepted by page 1, (2) avoid rejection with the rest. download both PDFs too.

maxwellforbes.com/posts/how-to...

11.04.2025 21:35 πŸ‘ 19 πŸ” 4 πŸ’¬ 1 πŸ“Œ 1

There’s no such thing as A.I β€œart.” It’s a soulless amalgamation of plagiarised data. That is the antithesis of art.

29.03.2025 12:48 πŸ‘ 8950 πŸ” 1545 πŸ’¬ 35 πŸ“Œ 2

I'm interested in doing a PhD in your group. Sent you an email, looking forward to your reply :)

29.03.2025 21:04 πŸ‘ 0 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0

#NeuroJobs

14.03.2025 23:56 πŸ‘ 0 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0
Preview
The Ultra-Scale Playbook - a Hugging Face Space by nanotron The ultimate guide to training LLM on large GPU Clusters

After 6+ months in the making and over a year of GPU compute, we're excited to release the "Ultra-Scale Playbook": hf.co/spaces/nanot...

A book to learn all about 5D parallelism, ZeRO, CUDA kernels, how/why overlap compute & coms with theory, motivation, interactive plots and 4000+ experiments!

19.02.2025 18:10 πŸ‘ 179 πŸ” 52 πŸ’¬ 2 πŸ“Œ 5
Preview
In your New Year’s resolutions for 2025, consider public outreach If every person in the neuroscience community committed to doing one thing, imagine the cumulative difference it would make.

Why & how scientists should do public outreach, especially now. Smart tips from @nicolecrust.bsky.social on @thetransmitter.bsky.social πŸ§ͺ www.thetransmitter.org/craft-and-ca...

04.01.2025 16:09 πŸ‘ 200 πŸ” 68 πŸ’¬ 8 πŸ“Œ 12