LabVeira @LabVeira

x.com/i/article/2088…

merinding banget nonton ini #INDONESIASADAR #DEMO

"Mathematical Theory of Deep Learning" is a free book that introduces the mathematical foundations of deep learning. The book focuses on three central areas of neural network theory: 1. approximation theory, which studies how neural networks represent functions; 2. optimization theory, which examines how their parameters are learned; 3. statistical learning theory, which investigates how models generalize beyond their training data. I think it's a very useful resource for anyone interested in studying deep learning from a mathematical perspective. arxiv.org/pdf/2407.18384

4 Awesome Books Everyone Should Read Atleast Once 📖📚

keyboard_arrow_left Previous keyboard_arrow_right Next

Serasa jadi detektif tiap habis baca ini 🕵️‍♀️ umur segini emg paling pas kalau harinya diisi dgn memecahkan misteri wkwk. 🧩 Murdoku Vol.1 (10/10)- addicting! 🧩 Murdle Vol.1 (9/10)- skill deduksi tinggi 🧩 The Killer is Here Vol.1 (8/10) 🧩 Murdsearch Vol.1 (6/10) -buat pemula

keyboard_arrow_left Previous keyboard_arrow_right Next

强烈推荐斯坦福的这套关于 AI Agent 的课程，对于有志成为 agent engineer 的同学可以看看：cs329z.stanford.edu

NVIDIA's head of robotics just held a funeral for the two things most robot brains are built on today, VLAs and teleop, in a 20-minute talk: 0:30 summer 2016: Jensen hand-delivers the first DGX-1 to OpenAI, and Jim signs it as an intern, right next to Karpathy 3:30 why VLAs are really "LVAs": most of the parameters go to language, so they're great at nouns and bad at verbs 8:00 why teleop can't scale: 24 hours per robot per day at best, closer to 3 in practice 11:00 EgoScale: 21K hours of human first-person video, zero robot data in pre-training, and 4 hours of teleop, under 0.1% of the mix 12:30 a scaling law for robot dexterity, 6 years after the one for language models 17:00 his bet: robots pass the physical Turing test in 2–3 years after watching I realized - the next robot brain won't be trained by people in VR headsets, it'll learn from videos of people making breakfast full talk by Jim Fan (@DrJimFan): youtube.com/watch?v=3Y8aq_… the article below is how to build an Optimus-style brain at home

Martin @0x_meden

5 days ago

Tesla won't sell you Optimus or show you its brain. but every humanoid brain built in public uses the same 3 layers, and all 3 are free to build at home for $0 to $230. last night I got a humanoid waving on my laptop. here's the 8-week path to its brain x.com/i/article/2113…

5 15 80 94K 177

🚨 5 countries. 6 warnings. 1 week.. 🇯🇵 Japan — on November 2, the day its new Disaster Management Agency opens, the government asks ALL citizens to store food, water and toilet supplies for at least 3 days.. a full week if they can.. the reason is big earthquakes.. 🇬🇧 UK — the new "Be Ready" campaign tells families to keep 3 days of water, canned food, cash and a radio.. the radios are now SOLD OUT.. 🇺🇸 US — the Pentagon wants its bases on US soil to run at least 90 days with no outside supply.. their own power, water, fuel and comms.. the plan is called FORTRESS.. its spokesman says "Power is combat now".. 🇬🇧 UK again — the head of the British Army, General Sir Roly Walker, stood in front of his soldiers at Catterick and told them "our nation is going to need you".. 🇩🇪 Germany — every 18-year-old man gets a military service questionnaire.. skip it and the fine is up to €250.. the first fines have started.. 🇳🇴 Norway — a government press release, titled in plain words.. "3 things you must do to prepare for crisis and war".. and this is its Self-Preparedness Week.. and here's the part that should worry you.. one by one, each of these looks like normal news.. put them in one list and it feels very different.. all of it while the US and Iran are still on edge.. and the one that asks a whole country to stock up hasn't even started yet.. Follow me.. on November 2, open it again and see if the list got longer..

Andrej Karpathy’s 1-hour Stanford lecture explains AI engineering beautifully. 10% → LLM 30% → Prompt 50% → Agent 70% → Loop 100% → Graph The real power isn’t just prompting models. It’s building systems around them with context, memory, tools, and feedback loops. “Delete everything, keep Graph.” A must-watch for anyone building AI agents. 🚀

ML engineer interview in 2016: - Explain linear regression - What's overfitting - Train a model in sklearn - Get the offer ML engineer interview in 2026:

作为一名博士，现在越来越怀疑，自己到底还有什么独特的优势。 以前做研究，查文献、想方案、写代码、解决问题，基本都是自己古法手搓。 辛辛苦苦读了几年博士，拿到 PhD，心里多少有点小骄傲。毕竟受过系统的科研训练，觉得自己至少在研究能力、解决复杂问题方面有些优势。 但现在，这点小傲娇在 AI 面前越来越不值一提。 工作上，大多数事情都是 Claude 完成的，方案是它提的，问题是它分析的，解决办法也是它想出来的。 有时候，自己最大的作用，就用鼠标点击一下Claude推荐的选项，让它继续执行。 这么基础的工作，也不需要博士学位吧？ AI 不仅在改变博士的工作，也在改变博士的培养。学生用 AI 做研究，导师用 AI 指导学生。选题、文献、代码、论文，AI 几乎都在参与。 那以后学生拿到的 PhD，到底证明了什么？ 证明他有独立研究的能力，还是证明他比较会使用 AI？又或者，只是证明他愿意花几年时间拿个博士学位？ 过去，博士学位多少还有点稀缺性。虽然绝大多数博士也只是普通人，做的研究、发的paper毫无价值，但至少证明这个人能吃苦、坐得住，愿意花几年时间钻研一个问题。 现在有了 AI，这些优势都在逐渐消失。 有时候甚至觉得，我们这代博士，可能是最后一批还能对自己的 PhD 学位有点小骄傲的人，还有点 “古法科研” 的经历可以吹。 再过几年，博士学位的吹逼点在哪里？

Gue jadi Mau Mulai Baca Novel Lagi Setelah Baca Jurnal Ini Ada penelitian yang melibatkan 19 orang yang diminta untuk membaca novel selama 9 hari. Setelahnya orang-orang tersebut diperiksa otaknya dengan fMRI. Nah hasilnya ternyata menunjukkan: -Peningkatan konektivitas otak di area pemahaman cerita, perspective-taking (mengambil sudut pandang orang lain), dan embodied semantics (merasakan pengalaman tokoh secara sensorik) -Efek jangka panjang di area otak berupa konektivitas baru Penelitian ini menunjukkan bahwa membaca cerita panjang, termasuk novel, bisa membantu untuk membuat kondisi otak menjadi lebih baik. Bisa jadi salah satu strategi juga untuk melawan brain rot. Semoga bermanfaat! Sumber Ilmiah: -Short- and Long-Term Effects of a Novel on Connectivity in the Brain

alexei @alexeixbt

2 days ago

no offense but reading is literally the cure to brain rot and there's no work around to reading books.

39 611 6K 218K 2K

Pemerintah Jepang & UK meminta warganya untuk siap darurat 🇬🇧 INGGRIS: 3 HARI 🇯🇵 JEPANG: 3 HARI Dan mereka tidak sendirian: 🇧🇪 Belgia: 72 JAM 🇮🇪 Irlandia: "Bersiap Untuk Keadaan Darurat" 🇫🇷 Prancis: 72 JAM 🇩🇪 Jerman: 3 hingga 10 HARI 🇳🇱 Belanda: 72 JAM 🇵🇱 Polandia: 3 HARI 🇸🇪 Swedia: 1 MINGGU 🇳🇴 Norwegia: 1 MINGGU 🇫🇮 Finlandia: 72 JAM 🇩🇰 Denmark: 3 HARI 🇪🇪 Estonia: 1 MINGGU 🇱🇻 Latvia: 72 JAM 🇱🇹 Lithuania: 72 JAM 🇨🇭 Swiss: 1 MINGGU 🇹🇼 Taiwan: 7 HARI 🇪🇺 EU: 72 JAM 17 negara + Uni Eropa. Apa yang akan terjadi ? Apakah mereka tahu sesuatu ?

holy f*ck, Karpathy put an entire AI degree on GitHub for free 7 repos, 306,741 stars, and in the right order they take you from backprop to agents that run research overnight each repo is the next layer of the one before 1. micrograd -> how a neural net actually learns backprop in about 100 lines of Python ↳ github.com/karpathy/micro… 2. nn-zero-to-hero -> the video course his lecture notebooks, from one neuron to GPT ↳ github.com/karpathy/nn-ze… 3. minbpe -> how text turns into tokens the tokenizer every LLM starts with, in clean minimal code ↳ github.com/karpathy/minbpe 4. nanoGPT -> your own GPT the simplest repo to train and fine-tune one ↳ github.com/karpathy/nanoG… 5. llm.c -> the same training with nothing hidden raw C and CUDA, no PyTorch underneath ↳ github.com/karpathy/llm.c 6. nanochat -> a full chat model end to end GPT-2 level training cost ~$43,000 in 2019 here it's about $48 ↳ github.com/karpathy/nanoc… 7. autoresearch -> agents that do the research an agent edits the training code, trains for 5 minutes, keeps what improved and repeats all night ↳ github.com/karpathy/autor… train one GPT from scratch and you become the person who gets handed the AI project, along with the raise that comes with it the classroom half of the same degree, what Stanford actually teaches about AI, is below ↓

Mr. Buzzoni @Mr_Buzzoni

4 months ago

x.com/i/article/2065…

41 176 1K 2.5M 4K

x.com/i/article/2114…

Will GPT-6 Astra kill robotics startups? Yes. And no. Since Astra was launched, 10 robotics researchers around me either cancel their projects or completely reconsider it. But I think the discussion now is sentiment more than fundamentals. In this video, I covered: 1) what exactly does Astra change 2) why LLM works like a doom 3) Can and Can't for LLM for robotics 4) Emerging moats and opportunities for robotics startup Open to any discussion in the comment.

so far ini 6 fiksi MANTAP yang kubaca tahun ini

🐻 @heninurh

3 days ago

Beli buku apalagi yaaa?? Coba bacaan paling best kalian tahun ini apa guys? Fiksi yaaah. Mau nyontek 😆

72 43 383 63K 309

How to learn Physical Intelligence's robot brain without working at Physical Intelligence: PI was in talks this year to raise $1B at an $11B valuation, and its brain, π0, is open source three of its co-founders teach the ideas inside it on YouTube for $0: Sergey Levine, Chelsea Finn and Quan Vuong here's the order I'd watch them in: 1: copy a human first Levine's Berkeley CS 285, lecture 2 (Spring 2026), is supervised learning of behaviors: a robot copies human demos, then drifts off course after its first small mistake. that drift is why ALOHA and π0 predict a whole chunk of motion at once instead of one step at a time 2: learn the math π0 moves with Stanford AA203, lecture 15 (Spring 2026). an hour in, it explains flow matching, the same math π0 uses to turn a decision into smooth arm motion 3: make it better after the demos run out Chelsea Finn's Stanford CS224R: lecture 16 is RL for real robots, and lecture 17 is a guest lecture from Ashish Kumar, who led AI for Tesla Optimus 4: see the whole brain in one talk Quan Vuong's guest talk in ETH's Robot Learning course is about π0.7. 13 minutes in, it folds a shirt on a robot arm that had no folding data of its own then stop watching and start running it: - openpi, PI's own code and π0 weights, 14.4K stars on GitHub - LeRobot by Hugging Face, 28.5K stars, which fine-tunes π0 on your own demos - a $122 SO-101 arm to record those demos on main insight: the lab in talks at $11 billion is built on three ideas stacked on top of each other: copy a human, move smoothly with flow matching, and let a vision-language model decide what to do next all three are free lectures now. the order is the only part nobody hands you P.S. my Optimus-style robot brain guide below turns this order into an 8-week plan

Martin @0x_meden

5 days ago

Tesla won't sell you Optimus or show you its brain. but every humanoid brain built in public uses the same 3 layers, and all 3 are free to build at home for $0 to $230. last night I got a humanoid waving on my laptop. here's the 8-week path to its brain x.com/i/article/2113…

5 15 80 94K 177

16 buku favoritku sepanjang masa, semua dari 16 ini memiliki alasan personalnya sendiri masing-masing dan sudah ada terjemahan bahasa Indonesia. 1. Le Petit Prince 2. The Count of Monte Cristo 3. Crime and Punishment 4. The Grapes of Wrath

keyboard_arrow_left Previous keyboard_arrow_right Next

If you want to stay up to date on the latest AI and Machine Learning research, reading academic papers can help. But they can be a bit intimidating and hard to understand sometimes. In this course, you'll learn how to approach and understand the theory, math, and structure in academic papers. freecodecamp.org/news/understan…