🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Can we teach Transformers Causal Reasoning? We propose Axiomatic Framework, a new paradigm for training LMs. Our 67M-param model, trained from scratch on simple causal chains, outperforms billion-scale LLMs and rivals GPT-4 in inferring cause-effect relations over complex graphs

statistical inference:

UW–Madison's "Mathematical Techniques for Algorithm Analysis" Lecture Notes: pages.cs.wisc.edu/~cs809-1/lectu… Course Material: pages.cs.wisc.edu/~cs809-1/

Terence Tao's "Linear Algebra" lecture notes PDF: terrytao.wordpress.com/wp-content/upl…

Graph Data Modeling in #Python — Practical guide to curating, analyzing, & modeling data with graphs: packtpub.com/en-us/product/… from Packt Data Science & Machine Learning #ad — #DataScience #AI #DataScientist — 𝒦𝑒𝓎 𝐹𝑒𝒶𝓉𝓊𝓇𝑒𝓈: 🔵Transform relational data models into graph data model while

この本の写経は英文を書く勉強としてとてもいいと思います。欧州数学会出版のたった50ページの本です。著者名Jerzy Trzeciak で検索おすすめ

FREE FREE FREE 10 Python Books, Absolutely FREE

『数学ガール／リーマン予想』ついに登場。 数学青春物語、堂々の完結へ！ ただいま予約受付中！ （2025年8月刊行予定） ◆結城浩『数学ガール／リーマン予想』 amzn.to/4lMB9hq #数学ガール

みなさま応援ありがとうございます！😭刊行はまだ先になりますが、ご予約いただきますと「非常に大きな追い風」となりますので、よろしくお願いいたします🙇もちろんリポスト、いいね、引用ポストなども感謝です！

*Deep Learning is Not So Mysterious or Different* by Andrew Gordon Wilson Fantastic paper showing that many interesting phenomena (e.g., double descent) can be understood in the frameworks of PAC-Bayes and "soft inductive biases". Great visuals! 😍 arxiv.org/abs/2503.02113

THE MICROARCHITECTURE OF PIPELINED AND SUPERSCALAR COMPUTERS AMOS R. OMONDI

At #ICLR2025 , check out our poster today on forecasting chaos with foundation models. Yuanzhao Zhang 章元肇 Poster #43. Thu 24 Apr, 3 - 5:30 pm in Hall 3 + Hall 2B

"Modern Discrete Probability, An Essential Toolkit" by Sebastien Roch PDF: people.math.wisc.edu/~roch/mdp/roch…

11 FREE Books from MIT for Absolute Beginners - Machine Learning (ML) - Deep Learning (DL) - Reinforcement Learning (RL) - Artificial Intelligence (AI)

How do language models generalize from information they learn in-context vs. via finetuning? We show that in-context learning can generalize more flexibly, illustrating key differences in the inductive biases of these modes of learning — and ways to improve finetuning. Thread: 1/

You can also solve the classic Double Descent problem (Vallet et. al. 1989) just using Random Matrix Theory Here's an outline of a sketch of the solution. This does not require replicas, and, instead, using just the properties of the Marchenko-Pastur distribution and it's

Crypto & AI:

Statistical physics for LLMs. Happy with that description :-) The original tweet thread for my paper: x.com/IAmTimNguyen/s… And my Machine Learning Street Talk interview: youtube.com/watch?v=W485bz…

Tired of deployment headaches? 😩 Our FREE Cloudflare learning roadmap provides a clear and structured path to understanding and utilizing Cloudflare for your web applications. roadmap.sh/cloudflare