I calculated a poster-size Transformer model by hand✍️

< Choosing a Vision Backbone > your model’s backbone is its perspective pick ResNet, and it sees in edges pick a ViT, and it sees in patches the backbone decides how your model thinks here are some of the most practical backbones and when you should choose them, from the

11 FREE Books from MIT for Absolute Beginners - Machine Learning (ML) - Deep Learning (DL) - Reinforcement Learning (RL) - Artificial Intelligence (AI) To get: - 1. Follow (So I can DM you ) 2. Like & retweet 3. Reply " Send "

I'm deleting this soon because it's a legit cash-printing formula. 𝗣𝗮𝗶𝗱 𝗖𝗼𝘂𝗿𝘀𝗲 𝗙𝗥𝗘𝗘 (PART - 3) 1. Artificial Intelligence + Data Analyst 2. Machine Learning + Data Science 3. Cloud Computing + Web Development 4. Ethical Hacking + Hacking 5. Data Analytics + DSA

Yes, it turns out diffusion models can learn from feedback as effectively as language models do with RL! Tsinghua, NVIDIA, and Stanford introduced Diffusion Negative-aware FineTuning (DiffusionNFT), a new online reinforcement learning paradigm that finally makes RL practical for

At MIT, I learned about RNNs in my NLP class with Prof. Michael Collins. He built a model from my keystrokes to predict who I was. To me, it felt like a magic box. Years later, when I had to teach RNNs, I forced myself to go inside the box. ⬇️ Download: byhand.ai/rnn

I split my 12 hr CUDA course into sections. We'll cover: 1) the deep learning ecosystem 2) cuda setup/installation 3) gentle intro to gpus 4) writing your first kernels 5) kernel and system level profiling, atomics, and the cuda programming model 6) how and when to use

The Andrej Karpathy interview 0:00:00 – AGI is still a decade away 0:30:33 – LLM cognitive deficits 0:40:53 – RL is terrible 0:50:26 – How do humans learn? 1:07:13 – AGI will blend into 2% GDP growth 1:18:24 – ASI 1:33:38 – Evolution of intelligence & culture 1:43:43 - Why self

New Harvard paper shows training‑free sampling lets a base LLM rival reinforcement learning on reasoning. No training, dataset, or verifier. The method samples from a power distribution, which means reweighting full sequences the model already thinks are likely. That bias

Evolution of Deep Learning by Hand ✍️ As my tribute to Geoff Hinton's Nobel Prize, I drew this animation to illustrate the key idea behind Hinton's major contributions to deep learning over the years, with artistic liberty. ---- 100% original, made by hand ✍️ Join 40k readers

Stanford's ALL FREE Courses [2024 & 2025] ❯ CS230 - Deep Learning ❯ CS234 - Reinforcement Learning ❯ CS236 - Deep Generative Models ❯ CME295 - Transformers & LLMs ❯ CS336 - Language Model from Scratch ❯ CS224N - NLP with DL Find links inside:

Stanford just released a new course for this Fall: Transformers & Large Language Models by the Amidi brothers. Three videos are already available for free on YouTube. SYLLABUS: > Transformers (tokenization, embeddings, attention, architecture) > LLM foundations (MoEs, types of

Stanford packed 1.5 hours with everything you need to know about LLMs. It explains why scale beats architecture and data beats genius. The clearest crash course on how AI actually works. Save this for later.

My pleasure to come on Dwarkesh last week, I thought the questions and conversation were really good. I re-watched the pod just now too. First of all, yes I know, and I'm sorry that I speak so fast :). It's to my detriment because sometimes my speaking thread out-executes my

fantastic simple visualization of the self attention formula. this was one of the hardest things for me to deeply understand about LLMs. the formula seems easy. you can even memorize it fast. but to really get an intuition of what the Q,K,V represent and interact, that’s hard.

Had the surreal experience of telling a room full of computer vision researchers at the ICCV25 AC workshop why “computer vision researcher” won’t be a thing in 5 years 🌶️ Of course, this was an extreme stance to keep things lively in a fun debate setting but it echoed some of my

Practical Linear Algebra for #DataScience — From Core Concepts to Applications Using #Python — amzn.to/3WWJKR4 ———— #DataScientist #AI #ML #MachineLearning #Mathematics #LinearAlgebra #Coding

Here's a neural net optimization trick that leads to ~4x faster CPU to GPU transfers. Imagine an image classification task. - We define the network, load the data and transform it. - In the training loop, we transfer the data to the GPU and train. Here's the problem with this:

This is still the way I recommend most people start with machine learning: 1. Start with Python 2. Learn to use Google Colab 3. Take a Pandas tutorial 4. Then a Seaborn tutorial 5. Learn how to use Decision Trees 6. Finish Kaggle's "Intro to Machine Learning" 7. Solve the

Complete Advance DSA Resources in One Place👇 From beginner sheets to advanced problem-solving guides, logic-building notes, and real-world DSA applications , everything you need to master Data Structures & Algorithms is HERE! Perfect for coding prep & product-based interviews.