barelyai's profile picture. Documenting my AI/ML journey | Interests : Philosophy, Mathematics, Psychology|

Laven

@barelyai

Documenting my AI/ML journey | Interests : Philosophy, Mathematics, Psychology|

Laven reposted

We can now train AI inside the mind of another AI. 🤯 🌍 Our world model, Genie 3, imagines and generates new worlds on the fly. 🤖 Our embodied agent, Sima, is dropped in and learns to navigate them autonomously. The entire loop—from the environment to the action—is generated…


This was so godamn good, I repeat it's extremely good. Really cleared so many misconceptions I previously had! youtu.be/RQWpF2Gb-gU

barelyai's tweet card. But what is quantum computing? (Grover's Algorithm)

youtube.com

YouTube

But what is quantum computing? (Grover's Algorithm)


> Decent enough in ML/DL to aim for research internships next summer > Proofs + Real Analysis + Mathematical Statistics in Pure Math > Decent conceptual coverage of DSA + Computer Architecture > Decently Fit ! Bit unrealistic given where I stand but only time will tell ✌🏼

my technical goals for this yr- >become really good at machine learning research(solo and team) >get a decentish rank in codeforces other imp things will flow by itself when these are done correctly,also some more things to be done share ur core goals for this yr



NAND gates --> NOT & AND gates Byte + enabler --> Register n NOT gates + 2^n AND Gates --> Decoder RAM --> 256 Registers for memory storage locations + 1 Memory Address Register (MAR) to select one of the storage locations. Boom!! We got fully half of a computer in our hands 🔥

barelyai's tweet image. NAND gates --> NOT & AND gates
Byte + enabler --> Register
n NOT gates + 2^n AND Gates --> Decoder
RAM --> 256 Registers for memory storage locations + 1 Memory Address Register (MAR) to select one of the storage locations.

Boom!! We got fully half of a computer in our hands 🔥
barelyai's tweet image. NAND gates --> NOT & AND gates
Byte + enabler --> Register
n NOT gates + 2^n AND Gates --> Decoder
RAM --> 256 Registers for memory storage locations + 1 Memory Address Register (MAR) to select one of the storage locations.

Boom!! We got fully half of a computer in our hands 🔥

Learnt how a combination of 4 NAND gates can remember & construct a memory bit. This is really fun ngl ✨

barelyai's tweet image. Learnt how a combination of 4 NAND gates can remember & construct a memory bit. This is really fun ngl ✨

Laven reposted

helix architecture is a 7B VLM and a 80M parameter encoder-decoder transformer that’s it wow

maharshii's tweet image. helix architecture is a 7B VLM and a 80M parameter encoder-decoder transformer

that’s it wow
maharshii's tweet image. helix architecture is a 7B VLM and a 80M parameter encoder-decoder transformer

that’s it wow
maharshii's tweet image. helix architecture is a 7B VLM and a 80M parameter encoder-decoder transformer

that’s it wow

Meet Helix, our in-house AI that reasons like a human Robotics won't get to the home without a step change in capabilities Our robots can now handle virtually any household item:



Crazy times to be alive, checkout this video to understand more about the recent breakthrough by Microsoft team in Quantum Computing by developing Majorana 1 chip in retrospect from 9 months ago!! youtu.be/ihZXl33t8So

barelyai's tweet card. Microsoft's Topological Quantum Computer Explained

youtube.com

YouTube

Microsoft's Topological Quantum Computer Explained


Richard Hamming was such a visionary!! And it's crazy to see how we've been able to maintain a straight line logarithmic growth curve (net effect) while avoiding the inevitable flattening of the S-curve of applications for so long 🤯 & imo we're doing it right now as well.

barelyai's tweet image. Richard Hamming was such a visionary!!

And it's crazy to see how we've been able to maintain a straight line logarithmic growth curve (net effect) while avoiding the inevitable flattening of the S-curve of applications for so long 🤯

& imo we're doing it right now as well.

China cooking on another level at this point 🔥

China is on 🔥 ByteDance drops another banger AI paper! OmniHuman-1 can generate realistic human videos at any aspect ratio and body proportion using just a single image and audio. This is the best i have seen so far. 10 incredible examples and the research paper Link👇



Wait... Why am I watching this at 12am, enjoyed it nevertheless ✨✨ youtu.be/HjneAhCy2N4

barelyai's tweet card. HOW TRANSISTORS RUN CODE?

youtube.com

YouTube

HOW TRANSISTORS RUN CODE?


Day 3 : - Bisection Search and Newton Raphson Method for finding polynomial roots (Lecture 6 - MIT 6.100L). - 3 Chapters of Introduction to Computation & Programming book to cover the readings till now. - 2 Chapters (till Chapter 4) of The Art of Doing Science & Engineering. ✌️

barelyai's tweet image. Day 3 :

- Bisection Search and Newton Raphson Method for finding polynomial roots (Lecture 6 - MIT 6.100L).
- 3 Chapters of Introduction to Computation & Programming book to cover the readings till now.
- 2 Chapters (till Chapter 4) of The Art of Doing Science & Engineering.

✌️
barelyai's tweet image. Day 3 :

- Bisection Search and Newton Raphson Method for finding polynomial roots (Lecture 6 - MIT 6.100L).
- 3 Chapters of Introduction to Computation & Programming book to cover the readings till now.
- 2 Chapters (till Chapter 4) of The Art of Doing Science & Engineering.

✌️
barelyai's tweet image. Day 3 :

- Bisection Search and Newton Raphson Method for finding polynomial roots (Lecture 6 - MIT 6.100L).
- 3 Chapters of Introduction to Computation & Programming book to cover the readings till now.
- 2 Chapters (till Chapter 4) of The Art of Doing Science & Engineering.

✌️
barelyai's tweet image. Day 3 :

- Bisection Search and Newton Raphson Method for finding polynomial roots (Lecture 6 - MIT 6.100L).
- 3 Chapters of Introduction to Computation & Programming book to cover the readings till now.
- 2 Chapters (till Chapter 4) of The Art of Doing Science & Engineering.

✌️

What a crazy day 😳😵‍💫

barelyai's tweet image. What a crazy day 😳😵‍💫

It's surreal to see how programming languages were built in the early days of computing (1950s), this one is an example of what Richard Hamming envisioned the design of his own programming language (L2) on his IBM 650 (programmed by him as well).

barelyai's tweet image. It's surreal to see how programming languages were built in the early days of computing (1950s), this one is an example of what Richard Hamming envisioned the design of his own programming language (L2) on his IBM 650 (programmed by him as well).

Day 2 : - Lecture 4 & 5 of MIT 6.100L + some problems. - 2 sections of Proofs book from CMU (till Obvious Obfuscation) - Russell's Paradox, Zermelo Fraenkel Set Theory's Axioms, Notion of Class, Function as set & way of generating new sets, Image Factorization of Functions. ✌️

barelyai's tweet image. Day 2 :

- Lecture 4 & 5 of MIT 6.100L + some problems.
- 2 sections of Proofs book from CMU (till Obvious Obfuscation)
- Russell's Paradox, Zermelo Fraenkel Set Theory's Axioms, Notion of Class, Function as set & way of generating new sets, Image Factorization of Functions. 

✌️
barelyai's tweet image. Day 2 :

- Lecture 4 & 5 of MIT 6.100L + some problems.
- 2 sections of Proofs book from CMU (till Obvious Obfuscation)
- Russell's Paradox, Zermelo Fraenkel Set Theory's Axioms, Notion of Class, Function as set & way of generating new sets, Image Factorization of Functions. 

✌️
barelyai's tweet image. Day 2 :

- Lecture 4 & 5 of MIT 6.100L + some problems.
- 2 sections of Proofs book from CMU (till Obvious Obfuscation)
- Russell's Paradox, Zermelo Fraenkel Set Theory's Axioms, Notion of Class, Function as set & way of generating new sets, Image Factorization of Functions. 

✌️
barelyai's tweet image. Day 2 :

- Lecture 4 & 5 of MIT 6.100L + some problems.
- 2 sections of Proofs book from CMU (till Obvious Obfuscation)
- Russell's Paradox, Zermelo Fraenkel Set Theory's Axioms, Notion of Class, Function as set & way of generating new sets, Image Factorization of Functions. 

✌️

Day 1 : - Speedran 3 lectures of MIT 6.100L + Psets. - Programming problems (10-12) to refresh the basics. - Calculus up till Mean Value Theoram. - Set Theory and Mathematical Logic (Münchhausen's Trilemma, Hilbert's Hotel, Zermelo Fraenkel set theory) ✌️

barelyai's tweet image. Day 1 :

- Speedran 3 lectures of MIT 6.100L + Psets.  
- Programming problems (10-12) to refresh the basics.
- Calculus up till Mean Value Theoram. 
- Set Theory and Mathematical Logic (Münchhausen's Trilemma, Hilbert's Hotel, Zermelo Fraenkel set theory)

✌️
barelyai's tweet image. Day 1 :

- Speedran 3 lectures of MIT 6.100L + Psets.  
- Programming problems (10-12) to refresh the basics.
- Calculus up till Mean Value Theoram. 
- Set Theory and Mathematical Logic (Münchhausen's Trilemma, Hilbert's Hotel, Zermelo Fraenkel set theory)

✌️
barelyai's tweet image. Day 1 :

- Speedran 3 lectures of MIT 6.100L + Psets.  
- Programming problems (10-12) to refresh the basics.
- Calculus up till Mean Value Theoram. 
- Set Theory and Mathematical Logic (Münchhausen's Trilemma, Hilbert's Hotel, Zermelo Fraenkel set theory)

✌️
barelyai's tweet image. Day 1 :

- Speedran 3 lectures of MIT 6.100L + Psets.  
- Programming problems (10-12) to refresh the basics.
- Calculus up till Mean Value Theoram. 
- Set Theory and Mathematical Logic (Münchhausen's Trilemma, Hilbert's Hotel, Zermelo Fraenkel set theory)

✌️

This was soul crushing to read as someone who aspires to be in that position, RIP & hope we can make AI space more kind & compassionate place to be for everyone!

This is the most gut-wrenching blog I've read, because it's so real and so close to heart. The author is no longer with us. I'm in tears. AI is not supposed to be 200B weights of stress and pain. It used to be a place of coffee-infused eureka moments, of exciting late-night arxiv…

DrJimFan's tweet image. This is the most gut-wrenching blog I've read, because it's so real and so close to heart. The author is no longer with us. I'm in tears. AI is not supposed to be 200B weights of stress and pain. It used to be a place of coffee-infused eureka moments, of exciting late-night arxiv…


Late to the trend, will start reading this ASAP.

barelyai's tweet image. Late to the trend, will start reading this ASAP.

Starting this classic right away!! It's been sitting in my reading list for so long, finally starting it before the new year starts. #philosophy

barelyai's tweet image. Starting this classic right away!! It's been sitting in my reading list for so long, finally starting it before the new year starts.

#philosophy

Finished! And all I could think after completing it was why I wasn't aware of this 5-7 years earlier, such great insights & perspective on learning anything in general.

Reading this right now & I've never had this kind of insight honestly about mind & learning techniques in general. Just read about 30% & already loving it !!

barelyai's tweet image. Reading this right now & I've never had this kind of insight honestly about mind & learning techniques in general. Just read about 30% & already loving it !!


Loading...

Something went wrong.


Something went wrong.