More or less specific resources for machine learning which I may or may not keep up to date. For general or theoretical machine learning resources, see:
Using machine learning libraries
Best known: PyTorch, TensorFlow
PyTorch
… can be tricky to install so here are some potential starting points with tips:
pip install light-the-torch
ltt install torch
Aggregated data on ML applications
Model development: training, evaluation
Domain-specific evals: code
- TL;DR 2025.04.14: LLMs are garbage in languages that aren’t Python
- mostly garbage at Java
- complete trash at Go, Rust, C, C++, Javascript, Typescript
Synthetic data
Synthetic data for privacy-preserving clinical risk prediction | Scientific Reports
Inference: using pretrained base models
Fine tuning IRL
[2312.05934] Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMs
Inference: using deployed models (agents, bots, apps)
Agents
Recent reasoning research: GRPO tweaks, base model RL, and data curation
DeepSeek's open-source week and why it's a big deal | PySpur - AI Agent Builder
Prompt engineering
Prompt Engineering Guide | Prompt Engineering Guide
(Twisted) RAG
Roaming RAG – RAG without the Vector Database - Arcturus Labs
Synthetic data
Synthetic data for privacy-preserving clinical risk prediction | Scientific Reports
[2407.01490] LLM See, LLM Do: Guiding Data Generation to Target Non-Differentiable Objectives
[2408.14960] Multilingual Arbitrage: Optimizing Data Pools to Accelerate Multilingual Progress
Data: ETL applications of AI
‘Learnable’ tasks (aka can AI/ML help?)
[2210.17011] A picture of the space of typical learnable tasks
Practical risks
Slopsquatting
Risky Bulletin: AI slopsquatting... it's coming!
The Rise of Slopsquatting: How AI Hallucinations Are Fueling...
Etc
Training an LLM from scratch for personal use
Not really something an individual would be expected to be able to do? (And even just fine tuning is hard!) Though, if you have the compute, it can be attempted, it takes a lot of time and effort, and it’s probably not going to be great, though you can try a light pretrain on domain specific data with fine tune on instructions to maybe get okay one shot performance.
And here’s a full screen recording of someone training a llama.cpp mini-ggml-model from scratch with the script to train.
Model distillation
The unreasonable effectiveness of reasoning distillation, Bespoke Labs
ML for science
The AI Scientist Generates its First Peer-Reviewed Scientific Publication
- paper.pdf
- SakanaAI/AI-Scientist-v2: The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search
Severe deviation in protein fold prediction by advanced AI: a case study | Scientific Reports
[2412.21154] Aviary: training language agents on challenging scientific tasks
How we evaluated Elicit Systematic Review
AI skepticism
Who and What comprise AI Skepticism? - by Benjamin Riley
Timelines
AI timelines: What do experts in artificial intelligence expect for the future? - Our World in Data
The History of Artificial Intelligence: Complete AI Timeline
Timeline of artificial intelligence - Wikipedia
The History of AI: A Timeline of Artificial Intelligence | Coursera
My AI Timelines Have Sped Up (Again)
The Timeline of Artificial Intelligence - From the 1940s to the 2020s
Google AI - Our AI journey and milestones
Literature Review of Transformative Artificial Intelligence Timelines | Epoch AI
AI Timeline: Key Events in Artificial Intelligence from 1950-2025
Timelines to Transformative AI: an investigation — EA Forum
Timeline of AI timelines - Timelines
AI Timeline Surveys – AI Impacts
The A.I. Timeline is Accelerating... - YouTube
Rising Tide | Helen Toner | Substack
Evaluating Large Language Models | Center for Security and Emerging Technology