Number 12 of 12 · November 18, 2026

Deep Learning for Physicists: From Linear Regression to Reasoning Models

These 115-page lecture notes by physicist Masanori Hanada teach deep learning, from linear regression to reasoning models, using physical intuition instead of heavy formalism. They claim no new results.

What they found

The paper is a set of lecture notes, not a research study. The author taught the course at Queen Mary University of London in 2023/24 and 2024/25. The notes begin with maximum likelihood, least squares, linear regression and gradient descent. They then cover neural networks, backpropagation, normalization, overfitting, adversarial examples and the kernel method. The second half covers language: word embeddings, recurrent networks, LSTMs, attention and Transformers. It ends with large language models, including scaling laws, instruction tuning, reinforcement learning from human feedback, hallucination, alignment faking, chain-of-thought reasoning and test-time scaling. The author says no originality is claimed and that developments after spring 2025 are not covered.

Why it matters

Many physicists and mathematicians want to understand modern SI but find the usual material hard to approach. These notes offer a self-contained path that assumes very little, and they even explain matrix multiplication from scratch. The abstract does not report any test of how well the notes teach. It only says readers with a physics or maths background may find them friendly.

How it changes the race

It does not change the race. It is teaching material with no new method or result. At most, it could make it easier for people from physics and maths to enter SI research.

Safety and control

The notes cover alignment faking and hallucination as topics in their large language model section. The abstract gives no detail on what they say about them.

The abstract, by the authors

These lecture notes are based on an introductory course on neural networks and deep learning that the author taught in the School of Mathematical Sciences at Queen Mary University of London in 2023/24 and 2024/25. The author, however, is a physicist, and it shows: wherever possible, ideas are explained with physical intuition rather than heavy formalism. Readers with a background in physics or mathematics who want a self-contained, introductory route into modern machine learning may therefore find these notes particularly friendly. The notes start from maximum likelihood, least squares, linear regression and (stochastic) gradient descent. They then cover neural networks, backpropagation, the vanishing gradient problem, batch and layer normalization, overfitting and regularization, adversarial examples, and the kernel method. The second half turns to natural language processing: word embeddings, recurrent networks and LSTMs, the attention mechanism and Transformers, and large language models, including scaling laws, the history of GPT, instruction tuning, reinforcement learning from human feedback, hallucination, alignment faking, chain-of-thought reasoning and test-time scaling. The emphasis on large language models reflects requests from the students. The notes are aimed at third- and fourth-year students in the UK system. Because the course was also taken by students from other departments, very little is assumed: even basic material such as matrix multiplication is explained from scratch, so readers familiar with it can safely skip such parts. No originality is claimed; the aim is a coherent, intuitive path through the material. Developments after spring 2025 are not covered, but the basic ideas emphasized here are unlikely to go out of date soon.

Quoted from arXiv:2611.21295. The research and its words belong to its authors.

The authors

Masanori Hanada

Credit for this research belongs to its authors. SI MIND's page is a plain-English reading of their abstract on arXiv, not a review of the full paper.