MOHIT
@dnn_code
undergrad • machine learning • life sciences • computational biology
Hii everyone, I'm a third-year BTech student. Interested in Machine Learning core, research, its application to physical domains and life sciences. Starting this twitter account to keep myself accountable, be honest with myself and keep a track.
How to win in your 20s For all those in their twenties, and those who want to reminisce about their twenties, have a read below: 1 Move to a big city. Don’t have to live there forever, but being in NYC or SF even for a short while will pay you back like crazy in your 30s. 2 Be…
Children under 5 are particularly vulnerable, about 1 in 5 AMR-attributable deaths are in young children. We need better prevention, vaccines, and early detection. LInk: ox.ac.uk/news/2022-01-2…
My PhD thesis--On Zero-Shot Reinforcement Learning--is now on arXiv.
vision RL ideas: -physics prediction. write a simulation consisting of a ball, some ramps/obstacles, and buckets beneath them (i.e. VPCT), VLM predicts which bucket the ball will fall into. reward for label correctness. -image editing. give a VLM access to photoshop, input…
Fundamentally intelligence is about: 1 - Ability to map the world or problem into an abstract navigable, updatable & flexible representation. 2 - Extract generalizable rules that govern a given space given few examples. 3 - Solve novel problems by navigating that space.
Fei-Fei Li (@drfeifei) on limitations of LLMs. "There's no language out there in nature. You don't go out in nature and there's words written in the sky for you.. There is a 3D world that follows laws of physics." Language is purely generated signal.
Way too many people think that AlphaFold "solved" ML for proteins. It didn't. It did revolutionize protein structure prediction, but that’s just one part of a much bigger puzzle. This is Part 1 of a series on what AlphaFold did (and didn’t) solve—and what comes next. ⬇️
Tough day today. 9 to 5 classes, pushed through 3 hours of sleep last night. After that studied for GATE a bit, and then explored ml for battery chemistry prediction, molecular dynamics, universal potentials etc.
so i got 30 people in a room yesterday on a sunday afternoon, and we read thru the book by @iskyzh where you write a Qwen2 serving model from scratch on MLX! skyzh.github.io/tiny-llm/
Working on two blog posts for next 2 weeks: 1. dna foundational models explained and literature review 2. protein language models, review, applications and downstream tasks Hopefully will complete them this time, instead of abandoning them.
For those looking to explore bio-ml and needing a beginning, following is a really great place to start. youtube.com/playlist?list=…
United States Trends
- 1. #BUNCHITA 1,407 posts
- 2. #SmackDown 46.1K posts
- 3. Tulane 4,358 posts
- 4. Frankenstein 79.8K posts
- 5. Giulia 15K posts
- 6. Aaron Gordon 4,222 posts
- 7. Supreme Court 184K posts
- 8. taylor york 8,200 posts
- 9. #TheLastDriveIn 3,952 posts
- 10. Russ 14K posts
- 11. #TheFutureIsTeal N/A
- 12. Justice Jackson 5,753 posts
- 13. Connor Bedard 3,061 posts
- 14. Gozyuger 1,697 posts
- 15. Podz 3,091 posts
- 16. Scott Frost N/A
- 17. #Blackhawks 1,787 posts
- 18. Northwestern 5,083 posts
- 19. Pluribus 31K posts
- 20. Tatis 2,130 posts
Something went wrong.
Something went wrong.