Machine Learning and AI Resources

“The goal is to turn data into information, and information into insight.”
~ Carly Fiorina

Machine learning and artificial intelligence have revolutionized the way we approach problem-solving in many fields, from healthcare to robotics to natural language processing. If you’re looking to deepen your understanding of these technologies, here are some of the best online resources and courses available.

Traditional Machine Learning

1. Micro, Macro & Weighted Averages of F1 Score, Clearly Explained

Computer Vision

1. Convolution in CNNs

Recurrent Neural Networks

1. Understanding LSTMs - Chris Olah
2. The Unreasonable effectiveness of Recurrent Neural Networks
3. Backpropogation through time - Mathematical Derivation
4. Why LSTMs Stop Your Gradients From Vanishing: A View from the Backwards Pass

Learning Methods

1. Contrastive Representation Learning

Famous Papers

1. OpenAI’s CLIP Paper - Explanation

Machine Learning & Deep Learning Courses

1. Practical Machine Learning with TensorFlow
   Learn to build machine learning models using TensorFlow.

2. Mathematics for Machine Learning
   A deep dive into the mathematical concepts that underpin machine learning algorithms.

3. Advanced Matrix Theory and Linear Algebra for Engineers
   Understand matrix theory and linear algebra with an emphasis on engineering applications.

4. Matrix Theory
   Learn the foundations of matrix theory, crucial for deep learning and machine learning algorithms.

5. Essential Mathematics for Machine Learning
   Build a strong mathematical foundation for machine learning and AI.

6. Machine Learning and Deep Learning Fundamentals
   This course provides a comprehensive introduction to machine learning and deep learning concepts.

7. Machine Learning
   A fundamental course to kickstart your journey in machine learning.

8. Machine Learning for Engineering and Science Applications
   Learn how machine learning is applied in engineering and scientific research.

9. Machine Learning And Deep Learning – Fundamentals and Applications
   A blend of theory and practical applications in machine learning and deep learning.

10. Deep Learning - Part 1
    Introduction to deep learning fundamentals, including neural networks and optimization techniques.

11. Deep Learning - Part 2
    Dive deeper into advanced deep learning concepts, architectures, and frameworks.

Natural Language Processing (NLP)

1. Natural Language Processing
   Learn the fundamentals of NLP, including text processing and feature extraction.

2. Natural Language Processing
   A comprehensive NLP course exploring algorithms and applications.

3. Applied Natural Language Processing
   Learn how to apply NLP techniques in real-world projects.

4. Deep Learning for Computer Vision
   Explore how deep learning models are applied to computer vision problems.

5. Deep Learning for Visual Computing
   Understand the intersection of deep learning and visual computing.

6. Introduction to Large Language Models - Tanmoy Chakraborty
   A course dedicated to large language models and their applications in NLP.

7. Introduction to Large Language Models - Mitesh Khapra
   Learn about large language models from an industry expert.

Reinforcement Learning & AI

1. Distributed Optimization and Machine Learning
   Explore the optimization techniques used in distributed machine learning systems.

2. Bandit Algorithm
   Learn the fundamentals of multi-armed bandit algorithms, useful in reinforcement learning.

3. Deep Generative Models
   Delve into the theory and applications of generative models like GANs and VAEs.

4. Reinforcement Learning
   An introduction to reinforcement learning, where agents learn by interacting with the environment.

5. Artificial Intelligence: Knowledge Representation and Reasoning
   Learn how knowledge can be represented and reasoned within AI systems.

6. Artificial Intelligence Search Methods For Problem Solving
   Study search algorithms, essential for AI problem-solving.

7. Applied Accelerated Artificial Intelligence
   Learn how to speed up and apply AI techniques in various industries.

8. Artificial Intelligence
   A comprehensive introduction to the field of artificial intelligence.

9. Artificial Intelligence
   Learn AI concepts and techniques applicable in real-world problems.

10. Pattern Recognition
    Understand pattern recognition and its applications across diverse fields.

---

Credits - Ajay Shenoy

Large Language Models

LLM Reasoning Papers

• LM Post-Training: A Deep Dive into Reasoning
  • https://arxiv.org/pdf/2502.21321
• A Tutorial on LLM Reasoning: Relevant Methods behind ChatGPT o1
  • https://arxiv.org/html/2502.10867v1
• Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning
  • https://arxiv.org/html/2502.14768v1

LLM Research Blogs

• LLM Research Newsletter
  • https://www.llmsresearch.com/
• Transformer Circuit thread
  • https://transformer-circuits.pub/

Internals

• Pytorch Internals
  • https://blog.ezyang.com/2019/05/pytorch-internals/
• Transformer Internals
  • https://goyalpramod.github.io/blogs/Transformers_laid_out/#coding-the-transformer

ICLR BLog Posts

• A New Alchemy: Language Model Development as a Subfield?
  • https://iclr-blogposts.github.io/2024/blog/language-model-development-as-a-new-subfield/
• Fairness in AI: two philosophies or just one?
  • https://iclr-blogposts.github.io/2024/blog/fairness-ai-two-phil-or-just-one/

Generative AI

1. Common pitfalls when building generative AI applications by Chip Huyen

2. ML CMU Blog

Mechanistic Interpretibility (MI)

Mechanistic interpretability aims to reverse-engineer a neural network into human-understandable mechanisms. MI focuses on transformers (specifically LLMs) but is not limited to these neural network architectures

People

1. Neel Nanda
2. Alignment Forum

Primer on LLMs

1. Large language models, explained with a minimum of math and jargon

Transformers

1. Transformers Interpretibility
2. 200 Concrete Open Problems in MI

Quick Guides to MI

1. What is Mechanistic Interpretability and where did it come from?
2. Introduction to Mechanistic Interpretability
3. “Mechanistic interpretability” for LLMs, explained

How to get started with MI ?

1. Concrete Steps to Get Started in Transformer Mechanistic Interpretability

Relevant Papers

1. Towards a Mechanistic Interpretation of Multi-Step Reasoning Capabilities of Language Models
2. A Practical Review of Mechanistic Interpretability for Transformer-Based Language Models
3. Mechanistic Interpretability for AI Safety : A Review

Straight from Anthropic

1. Mapping the mind of a Large Language model
2. Interpretibility Dreams
3. Golden Gate Claude
4. Toy Models of Superposition
5. Transformer Circuits Thread

Blogs

1. Neel Nanda’s case on why we need interpretibility research
2. A Microscope into the Dark Matter of Interpretability

Libraries

1. Transfomer Lens
2. Neuronpedia
3. Interpretable Neural Networks

Why we need MI Research ?

Neel Nanda makes a couple of strong arguments here (15 in fact!) on why interpretibility research is needed and how it will help us resolve x-issues