Siraj Raval

@sirajraval

On a mission to help eliminate human disease by the end of this century. AI Educator | Youtuber | Author | Serial Entrepreneur | Founded

San Francisco, CA
Vrijeme pridruživanja: svibanj 2014.

Tweetovi

Blokirali ste korisnika/cu @sirajraval

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @sirajraval

  1. prije 17 sati

    I'm enjoying browsing the CUDA docs to learn useful GPU programming functions for a project. It's easy to build neural nets in Python, but to get real algorithmic speedups, we've got to get closer to the C/C++ core of these libraries. I have tutorials on CUDA dropping next week!

    Poništi
  2. 3. velj

    I'm beyond hype that DeepMind released the code, model, & dataset for their SOTA protein folding system titled AlphaFold! This is SUCH a big deal and in this episode, I'll explain how it works + installation steps to get you started.

    Poništi
  3. 2. velj

    I have a deep love for the AI community because whenever I've felt lost while trying to learn a difficult concept, there was always someone willing to help guide me. Feeling healthy, happy, and grateful to be alive!

    Poništi
  4. 1. velj

    This blog has the highest quality Quantum Computing explanations I've ever seen & it uses spaced-repetition to help you memorize important concepts. Now that's innovative education, shoutout to Michael Nielsen!

    Poništi
  5. 31. sij

    DeepMind just released the code, model, & dataset behind their groundbreaking "AlphaFold" system. It's predicts protein shapes from genomic data with apps in health, sustainability, & materials design. My video tutorial on how it works & how to install/run it drops next week!

    Poništi
  6. 30. sij

    There are 43K suicides every year in the USA, and few ever received mental health treatment. Our health system is so fixated on saving people from their inevitable death that it marginalizes the anguish of human existence. Psilocybin is a powerful therapeutic tool that can help

    Poništi
  7. 29. sij

    If you're interested in Machine Learning but don't feel you have the necessary math background, this newly released free book is the perfect primer to get you up to speed! I love this so much

    Poništi
  8. 28. sij

    Building powerful Computer Vision apps has never been easier than it is today, but why does a model make a certain image classification? Grad-CAM is a technique that helps shed some light, I'll explain it in this episode

    Poništi
  9. 27. sij

    Google's newly released Reformer architecture makes state of the art Natural Language Processing accessible on a budget! In this episode, I'll explain it's novel features and we'll use it to analyze a Biology textbook

    Poništi
  10. 26. sij

    The best part of DeepMind's Graph Nets library is the fact that they have Colab notebooks ready to demo in your browser for 3 different applications (Path Finding, Sorting, and Physics State Prediction). Have a look!

    Poništi
  11. 25. sij

    Google recently released the "Reformer" in Tensorflow, a language model that can process massive text sequences on a single accelerator. Now, the most popular version of it on GitHub is written in PyTorch. A sign of the times!

    Poništi
  12. 24. sij

    Natural Language Processing paper review never gets old, it sits at the most delightful intersection of linguistics & mathematics. And I'm loving these Sesame Street language model naming conventions (BERT, ELMO, ERNIE). My video on the new Reformer architecture drops next week!

    Poništi
  13. 23. sij

    To use "Artificial Intelligence", you need algorithms, computing, data, & education. Although most data is locked away by various institutions, fortunately this site has 3667 free health datasets you can use for your next project, check it out!

    Poništi
  14. 23. sij

    “Wherever the mind wanders, restless & diffuse in its search for satisfaction without, lead it within; train it to rest in the Self. Abiding joy comes to those who still the mind. Freed from the taint of self-will, with consciousness unified, they become one”- Gita, on meditation

    Poništi
  15. 22. sij

    2 common Machine Learning terms are 'cross entropy' & 'KL-Divergence', both used for optimization. To date, this is still the best explanation i've ever seen of these concepts. Shoutout to Aurelien Geron!

    Poništi
  16. 21. sij

    The Pharmaceutical industry needs to be upgraded big time, and ML is the perfect tool to do that. In this episode, I'll explain how you can use Insilio Medicine's recently open sourced DeepRL library to generate novel drugs seamlessly.

    Poništi
  17. 20. sij

    Graph networks have been around for 10 years, but its only now that we're seeing them being applied to some really cool problems. I'll help you decide which library to use to build your own between PyTorch Geometric,Deep Graph, & Graph Nets.

    Poništi
  18. 19. sij

    Advances in generative modeling like DeepFakes & StyleGAN2 are going to become one-click solutions soon. When machine-generated internet personalities go mainstream, the human aspects of being a creator (humility, authenticity, novelty, vulnerability) will matter even more.

    Poništi
  19. 18. sij

    This web demo shows how photorealistic the outputs of the StyleGAN2 model are. . It's very accurately able to learn the input data distribution. I bet it could also be used as a rapid experimentation tool for drug discovery!

    Poništi
  20. 17. sij

    Have you ever encountered a research paper with a paywall? Just type the URL into Sci-Hub to get free access to it . Shoutout to Alexandra Elbakyan, the pirate Queen of academic papers for helping make research publicly available!

    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·