Background Readings

Here are a few resources for background readings, highly recommended for students both from computational and biological backgrounds:

1. Artificial Intelligence and Brain Research – Neural Networks, Deep Learning and the Future of Cognition:
   • https://link-springer-com.proxy2.library.illinois.edu/book/10.1007/978-3-662-68980-6 (pdf link: https://drive.google.com/file/d/1uOoUA-5XusmTJkrSMGWL0itMm5DCOLpr/view?usp=drive_link)
   • This book provides a nice and accessible introduction to both neuroscience (Part I, from neuron biology to consciousness and free will) and artificial intelligence (Part II, from artificial neurons to large language models).
2. Stanford CS 225: Machine Learning
   • https://cs229.stanford.edu/
   • Course materials with informative and detailed lecture notes on machine learning algorithms.
3. Molecular Biology for Computer Scientists
   • https://tandy.cs.illinois.edu/Hunter_MolecularBiology.pdf
   • A chapter from the book Artificial Intelligence and Molecular Biology that introduces biology for computer science backgrounds, including genetics and evolution.

More on Large Language Models (LLMs) and Transformer architectures:

1. Transformers, the tech behind LLMs | Deep Learning Chapter 5
   • https://www.youtube.com/watch?v=wjZofJX0v4M&list=PLZHQObOWTQDNU6R1_67000Dx_ZCJB-3pi&index=7
   • Part of a video series (from 3Blue1Brown) with great visualizations of the transformer architecture and the operations.
2. Build a Large Language Model (From Scratch)
   • https://i-share-uiu.primo.exlibrisgroup.com/view/action/uresolver.do?operation=resolveService&package_service_id=55001343200005899&institutionId=5899&customerId=5815&VE=true (need to log into O’Reily via UIUC netID)
   • A step-by-step book guide on building an LLM from scratch, with lots of code and explanations.
   • Also, see the HuggingFace transformer repository (https://github.com/huggingface/transformers/) for more code on building transformers.
3. Foundations of Large Language Models
   • https://arxiv.org/abs/2501.09223
   • Full book on foundational concepts of LLMs (with some more math), with 5 chapters exploring key areas: pre-training, generative models, prompting, alignment, and inference.
4. A Mathematical Framework for Transformer Circuits
   • https://transformer-circuits.pub/2021/framework/index.html
   • A blog paper from Anthropic on Transformer interpretability, highly recommended to everyone who is interested in the mathematical rationale behind the success of Transformers.
5. A few other (blog post) links: https://jalammar.github.io/illustrated-transformer/, https://e2eml.school/transformers.html.

More on Reinforcement Learning (RL):

1. Reinforcement Learning 101
   • https://towardsdatascience.com/reinforcement-learning-101-e24b50e1d292
   • A brief post from Medium that provides an introduction to RL
2. UC Berkeley CS 285: Deep Reinforcement Learning course, fall 2023
   • https://www.youtube.com/watch?v=tbLaFtYpWWU&list=PL_iWQOsE6TfVYGEGiAOMaOzzv41Jfm_Ps&index=4
   • This is a playlist for a lecture series that introduces reinforcement learning (RL) and goes more in-depth (including deep RL and language models). It is accessible to students without much computational background.
3. Reinforcement Learning: An Introduction (textbook by Sutton and Barto)
   • https://web.stanford.edu/class/psych209/Readings/SuttonBartoIPRLBook2ndEd.pdf
   • A classical introduction to RL, highly recommended.
4. Key papers in Deep RL
   • https://spinningup.openai.com/en/latest/spinningup/keypapers.html

More on Biology, Neuroscience, and Evolution:

1. Cognitive Biology: Dealing with Information from Bacteria to Minds
   • https://academic-oup-com.proxy2.library.illinois.edu/book/8903
   • A book that provides an intro to Cognitive Biology, looking at information flow at different scales in biological systems.