iiinova — where your digital self keeps evolving

Reposted by Duc Nguyen Huu

Gaël Varoquaux @gaelvaroquaux.bsky.social · 26/05/2026

In this high-level yet detailed blog post, I expend on the statistics that hold together AI and data science: blog.probabl.ai/data-science...

blog.probabl.ai

Data science is not AI, but it is its genesis and its new frontier

Gaël Varoquaux explains why and how statistical thinking is the heart of AI.

1136

Reposted by Duc Nguyen Huu

Kevin Markham @dataschool.io · 13/03/2026

My book is the #1 New Release in NLP! 🥳 Amazon US put it on sale... for $0.95 off 😂 Get the paperback: geni.us/MasterML Or read online (free!): mlbook.dataschool.io #MachineLearning #Python @scikit-learn.org

064

Reposted by Duc Nguyen Huu

Kevin Markham @dataschool.io · 27/02/2026

My new book - on sale NEXT WEEK! 🎉 Sign up to get notified when it's available: dataschool.kit.com/mlbook #MachineLearning #Python @scikit-learn.org

031

Duc Nguyen Huu @ducnh279.bsky.social · 20/02/2026

When I first started learning data science, I often got lost in the @scikit-learn.org documentation. After taking the course that became this book, I understood it much better and gained confidence using it. Scikit-learn is a powerful library,but without guidance, it can feel overwhelming at first.

110

Reposted by Duc Nguyen Huu

Kevin Markham @dataschool.io · 11/09/2025

Dream unlocked: I'm publishing my first book! 🎉🎉🎉 It's called "Master Machine Learning with scikit-learn: A Practical Guide to Building Better Models with Python" Download the first 3 chapters right now: 👉 dataschool.kit.com/mlbook 👈 Thanks for your support 🙏

1266

Reposted by Duc Nguyen Huu

David Holzmüller @dholzmueller.bsky.social · 29/07/2025

I got 3rd out of 691 in a tabular kaggle competition – with only neural networks! 🥉 My solution is short (48 LOC) and relatively general-purpose – I used skrub to preprocess string and date columns, and pytabkit to create an ensemble of RealMLP and TabM models. Link below👇

2112

Reposted by Duc Nguyen Huu

Gaël Varoquaux @gaelvaroquaux.bsky.social · 09/07/2025

This work is presented at ICML next week. • The paper arxiv.org/html/2502.05... • The python package: pypistats.org/packages/tab... (try it out 🐍) • The source code github.com/soda-inria/t... (100% open source, including pre-training 💞) Longer read (5mn): gael-varoquaux.info/science/tabi... 8/9

arxiv.org

TabICL: A Tabular Foundation Model for In-Context Learning on Large Data

0122

Reposted by Duc Nguyen Huu

Gaël Varoquaux @gaelvaroquaux.bsky.social · 09/07/2025

👨‍🎓🧾✨#icml2025 Paper: TabICL, A Tabular Foundation Model for In-Context Learning on Large Data With Jingang Qu, @dholzmueller.bsky.social, and Marine Le Morvan TL;DR: a well-designed architecture and pretraining gives best tabular learner, and more scalable On top, it's 100% open source 1/9

15115

Reposted by Duc Nguyen Huu

Kevin Markham @dataschool.io · 28/05/2025

My thoughts on the current state of AI progress and the most important developments in 2025: www.dataschool.io/ai-progress-...

dataschool.io

AI progress in 2025 📈

Thoughts on the current state of AI progress and the most important developments in 2025

011

Duc Nguyen Huu @ducnh279.bsky.social · 05/04/2025

Are you familiar with Token Pooling? Models that use late interaction, like ColBERT, ColPali, and ColQwen, gain significant benefits from this pooling technique! By integrating token pooling methods, the number of vectors to store can be reduced. Blog: www.answer.ai/posts/colber...

answer.ai

A little pooling goes a long way for multi-vector representations – Answer.AI

Practical AI R&D

000

Duc Nguyen Huu @ducnh279.bsky.social · 04/04/2025

Efficiently scale long CoT models like DeepSeek when using Best-of-N or Majority Voting by early pruning reasoning chains. Kaggle Discussion: www.kaggle.com/competitions...

kaggle.com

AI Mathematical Olympiad - Progress Prize 2

Solve national-level math challenges using artificial intelligence models

000

Duc Nguyen Huu @ducnh279.bsky.social · 31/03/2025

I find making your agents safe is just as important as making them smart. 🔒 A good read for building secure AI! arxiv.org/pdf/2503.18813

010

Duc Nguyen Huu @ducnh279.bsky.social · 30/03/2025

There will be one day ... in 🇺🇸 or 🇻🇳

210

Reposted by Duc Nguyen Huu

Kevin Markham @dataschool.io · 27/03/2025

Claude finally integrated web search into its results... But with LangChain & LangGraph, you can build a chatbot that integrates web search into ANY model you like! You'll learn how to do that (and much more) in my new AI course... Sign up for EARLY ACCESS: 👉 dataschool.kit.com/agents 👈

022

Duc Nguyen Huu @ducnh279.bsky.social · 24/03/2025

A practical way for students to secure jobs and earn money is by developing real-world projects. Researching or engineering LLMs often seems like a field dominated by the big tech! It's still important to learn fundamentals from scratch for growth and problem-solving (e.g be able to fix things)! 😁

020

Reposted by Duc Nguyen Huu

Sebastian Raschka (rasbt) @rasbt.bsky.social · 23/03/2025

My next tutorial on pretraining an LLM from scratch is now out. It starts with a step-by-step walkthrough of understanding, calculating, and optimizing the loss. After training, we update the text generation function with temperature scaling and top-k sampling: www.youtube.com/watch?v=Zar2...

06112

Reposted by Duc Nguyen Huu

Rashmi Banthia @rashmib.bsky.social · 20/03/2025

cuDF-pandas (%load_ext cudf.pandas) with Rapids ... work similarly and super cool to see we will be able to speed up scikit-learn

111

Reposted by Duc Nguyen Huu

Kevin Markham @dataschool.io · 17/03/2025

In honor of March Madness 🏀, I've got a new blog post: www.dataschool.io/pandas-strea... Learn how to identify & analyze scoring streaks using pandas operations: - shift() - cumsum() - boolean math - groupby()

dataschool.io

How to calculate "scoring streaks" with pandas 🏀

Learn how to identify & analyze consecutive events in your data using advanced DataFrame methods!

011

Duc Nguyen Huu @ducnh279.bsky.social · 18/03/2025

Many good advices/best practices for missing value imputation in the paper! I now have a much deeper appreciation for Data School's course and regard it as the best scikit-learn course. Master Machine Learning with scikit-learn: courses.dataschool.io/master-machi...

courses.dataschool.io

Course: Master Machine Learning with scikit-learn

Get ready for your dream job in Machine Learning!

121

Reposted by Duc Nguyen Huu

Kevin Markham @dataschool.io · 13/03/2025

"Some people today are discouraging others from learning programming on the grounds AI will automate it. This advice will be seen as some of the worst career advice ever given." -- Andrew Ng, legendary AI researcher Source: www.deeplearning.ai/the-batch/is...

deeplearning.ai

DeepSeek-R1 Uncensored, QwQ-32B Puts Reasoning in Smaller Model, and more...

The Batch AI News and Insights: Some people today are discouraging others from learning programming on the grounds AI will automate it.

011

Reposted by Duc Nguyen Huu

Gaël Varoquaux @gaelvaroquaux.bsky.social · 14/03/2025

A recent talk, fully in a vscode: 100% code on data wrangling for machine learning with @skrub-data.bsky.social www.youtube.com/watch?v=hdWW... super powerful to easily assemble production-ready pipelines in easy syntax

youtube.com

The Future of AI & Machine Learning | The Python Exchange February 2025

YouTube video by Don't Use This Code • James Powell

2235

Reposted by Duc Nguyen Huu

Sebastian Raschka (rasbt) @rasbt.bsky.social · 13/03/2025

Yesterday, Google released Gemma 3, their latest open-weight LLM. Finally, a new addition to the "Big 5" of open-weight models (Gemma, Llama, DeepSeek, Qwen, and Mistral). I just went through the Gemma 3 report and experimented a bit with the models, and there are plenty of interesting tidbits:

35811

Reposted by Duc Nguyen Huu

Sebastian Raschka (rasbt) @rasbt.bsky.social · 11/03/2025

Just uploaded my "Coding Attention Mechanisms" tutorial. A 2h15m session on coding attention mechanisms to understand how the engine of LLMs works: self-attention → parameterized self-attention → causal self-attention → multi-head self-attention www.youtube.com/watch?v=-Ll8...

youtube.com

Build an LLM from Scratch 3: Coding attention mechanisms

YouTube video by Sebastian Raschka

0364

Reposted by Duc Nguyen Huu

Sebastian Raschka (rasbt) @rasbt.bsky.social · 08/03/2025

I just shared a new article, "The State of Reasoning Models", where I am exploring 12 new research articles on improving the reasoning capabilities of LLMs (all published after the release of DeepSeek R1): magazine.sebastianraschka.com/p/state-of-l... Happy reading!

magazine.sebastianraschka.com

The State of LLM Reasoning Models

Part 1: Inference-Time Compute Scaling Methods

16114

Reposted by Duc Nguyen Huu

Trey Hunner @trey.io · 03/03/2025

A couple months ago @dataschool.io wrote about a tool he uses to chat with different LLM models without paying a monthly subscription to all of them. The tool is called Typing Mind and I decided to pay $30 for lifetime access. It was well worth it. Kevin's post 👇 www.dataschool.io/save-money-o...

dataschool.io

Use premium AI models for pennies 💰

Learn how to access ChatGPT, Claude, and more for pennies per conversation rather than paying for expensive subscriptions!

442

Reposted by Duc Nguyen Huu

Kevin Markham @dataschool.io · 04/03/2025

19 professionals (in a variety of fields) evaluated OpenAI's Deep Research vs Google's Deep Research. OpenAI was the clear winner 🏆 Neat study by @binarybits.bsky.social, read more here: www.understandingai.org/p/these-expe...

043

Reposted by Duc Nguyen Huu

Sebastian Raschka (rasbt) @rasbt.bsky.social · 02/03/2025

A new tutorial in my “Build A Large Language Model From Scratch” series is now live (www.youtube.com/watch?v=341R...) - Tokenizing raw text and converting tokens into token IDs - Applying byte pair encoding - Setting up data loaders in PyTorch for efficient training

youtube.com

Build an LLM from Scratch 2: Working with text data

YouTube video by Sebastian Raschka

1446

Duc Nguyen Huu @ducnh279.bsky.social · 20/02/2025

Another great read on reasoning models! 🧠 Small LMs struggle to learn from long or complex CoTs from larger teachers. 🔍 Why? The reasoning complexity may be too overwhelming. 🚀 Solution: Mix simple & complex CoTs! 📈 Results: Clear gains over complex CoTs alone! Arxiv: arxiv.org/pdf/2502.121...

arxiv.org

000

Duc Nguyen Huu @ducnh279.bsky.social · 19/02/2025

OMGGG!!!!!!!!!!! 100+ pages book to fully understand distributed training. Author: "🎯 Our goal: democratize knowledge about LLM training at scale. Whether you're starting with 1 GPU or orchestrating thousands, this guide walks you through the journey step by step." huggingface.co/spaces/nanot...

huggingface.co

The Ultra-Scale Playbook - a Hugging Face Space by nanotron

The ultimate guide to training LLM on large GPU Clusters

020

Reposted by Duc Nguyen Huu

Kevin Markham @dataschool.io · 16/02/2025

Here's my personal guide to Bluesky happiness 🌈 1. Be very selective about who I follow (42 people at the moment) 2. Read everything in the "Following" feed, and nothing else 3. Trust that the people I follow will surface other good people to follow 4. Unfollow anyone who no longer makes the cut

011

Reposted by Duc Nguyen Huu

Sebastian Raschka (rasbt) @rasbt.bsky.social · 15/02/2025

It's 2025, and I’ve finally updated my Python setup guide to use uv + venv instead of conda + pip! Here's my go-to recommendation for uv + venv in Python projects for faster installs, better dependency management: github.com/rasbt/LLMs-f... (Any additional suggestions?)

1115920

Reposted by Duc Nguyen Huu

Kevin Markham @dataschool.io · 12/02/2025

Wondering about the differences between "Jupyter", "Jupyter Notebook", "JupyterLab", "IPython", "Colab", and other related terms? I'll explain these terms (and more) in 5 minutes: ▶️ www.youtube.com/watch?v=TDlG...

youtube.com

Jupyter & IPython terminology explained

YouTube video by Data School

021

Duc Nguyen Huu @ducnh279.bsky.social · 12/02/2025

Reading this paper, you feel the power of RL. RL is great, but not all we need. A blend of RL, inference time scaling, good pretraining, and domain-specific tuning is often the way to go! arxiv.org/pdf/2502.06807

arxiv.org

000

Duc Nguyen Huu @ducnh279.bsky.social · 12/02/2025

LLM Trick - Finding the Near Optimal Temperature for Multi-Sample Inference (e.g., Majority Vote, Best-of-N) - Notebook (implement paper): 🤖https://github.com/ducnh279/Optimize-Temperature-for-LLM-Inference/blob/main/optimize-temperature-for-llm-inference.ipynb

110

Duc Nguyen Huu @ducnh279.bsky.social · 11/02/2025

When I first started learning about LLMs, I used this cheat sheet (community.openai.com/t/cheat-shee...) to choose sampling "temperature." Now, I’ve found another useful technique to add to my bag of tricks: arxiv.org/pdf/2502.05234.

bit.ly

Cheat Sheet: Mastering Temperature and Top_p in ChatGPT API

Hello everyone! Ok, I admit had help from OpenAi with this. But what I “helped” put together I think can greatly improve the results and costs of using OpenAi within your apps and plugins, specially ...

030

Duc Nguyen Huu @ducnh279.bsky.social · 11/02/2025

Have you ever wondered what are the differences between 2 types of knowledge distillation (for LLMs) these days? Link: developer.nvidia.com/blog/how-to-...

010

Duc Nguyen Huu @ducnh279.bsky.social · 10/02/2025

My learning project: Implement a Reasoning Model with GRPO from scratch Code: github.com/ducnh279/grp... - Base model: Qwen2.5-1.5B - Dataset: GSM8K (math) - Reward functions: Format, Accuracy - Objective: GRPO - PEFT: LoRA

150

Reposted by Duc Nguyen Huu

Sebastian Raschka (rasbt) @rasbt.bsky.social · 09/02/2025

Maybe a hot take, but what about the following advice to the next gen: Don't get an AI degree; the curriculum will be outdated before you graduate. Instead, study math, stats, or physics as your foundation, and stay current with AI through code-focused books, blogs, and papers.

1214722

Duc Nguyen Huu @ducnh279.bsky.social · 08/02/2025

𝐂𝐨𝐬𝐢𝐧𝐞 𝐫𝐞𝐰𝐚𝐫𝐝 stabilizes training & incentivizes the model to generate responses that are both accurate and reasonably lengthy. To prevent reward hacking, a repetition penalty is added to the cosine reward. Paper: arxiv.org/pdf/2502.03373 Hackable Code: github.com/eddycmu/demy...

010

Reposted by Duc Nguyen Huu

Kevin Markham @dataschool.io · 07/02/2025

Comparing: - Software Developers - ML Engineers - LLM Developers - Prompt Engineers Read here: www.louisbouchard.ai/llm-develope... Image & article by Louis-François Bouchard

011

Reposted by Duc Nguyen Huu

Sebastian Raschka (rasbt) @rasbt.bsky.social · 07/02/2025

Just read the s1: Simple Test-Time Scaling paper. Super interesting approach to improving reasoning models! TL;DR: 1. SFT on 1k curated examples w/ reasoning traces. 2. Control response length w/ budget forcing: "Wait" tokens → longer reasoning/self-correction. "Final Answer:" → enforce stopping.

2376

Reposted by Duc Nguyen Huu

Sebastian Raschka (rasbt) @rasbt.bsky.social · 05/02/2025

I just finished writing up my take on reasoning models: magazine.sebastianraschka.com/p/understand... Here, I 1. Discuss the advantages & disadvantages of reasoning models 2. Of course, describe and discuss DeepSeek R1 3. Describe the 4 main ways to building & improving reasoning models

magazine.sebastianraschka.com

Understanding Reasoning LLMs

Methods and Strategies for Building and Refining Reasoning Models

39020

Duc Nguyen Huu @ducnh279.bsky.social · 05/02/2025

The best way to understand GRPO is to code it from scratch without HF Trainer. 😂

100

Duc Nguyen Huu @ducnh279.bsky.social · 05/02/2025

Want to start with RL basics for training reasoning LLMs using GRPO? Check out: superb-makemake-3a4.notion.site/group-relati...

superb-makemake-3a4.notion.site

group relative policy optimization (GRPO) | Notion

intro

010

Reposted by Duc Nguyen Huu

Kevin Markham @dataschool.io · 31/01/2025

This blog post is now available in video form: www.youtube.com/watch?v=rOKX... (Hopefully I did a good job pronouncing Andrew Ng & Sebastian Raschka 🤷‍♂️)

youtube.com

How to keep up with AI in 2025

YouTube video by Data School

021

Reposted by Duc Nguyen Huu

Kevin Markham @dataschool.io · 30/01/2025

Want to keep up with the latest developments in AI? Check out the 7 newsletters I personally read & recommend for AI news, insights, and analysis: 👉 www.dataschool.io/best-ai-news... 👈 Includes: @simonwillison.net @emollick.bsky.social @sebastianraschka.com @binarybits.bsky.social

dataschool.io

How to keep up with AI in 2025 🏃‍♂️

These 7 AI experts will guide you through the most important developments in Artificial Intelligence.

3143

Reposted by Duc Nguyen Huu

Sebastian Raschka (rasbt) @rasbt.bsky.social · 24/01/2025

But before I get to the reasoning model space... if you are looking to do some focused offline reading this weekend, I just re-compiled my take on the "noteworthy AI research papers of 2024" into one PDF-export-friendly 47-page mega-post with TOC and all: sebastianraschka.com/blog/2025/ll...

sebastianraschka.com

Noteworthy LLM Research Papers of 2024

This article covers 12 influential AI research papers of 2024, ranging from mixture-of-experts models to new LLM scaling laws for precision.

56113