The best benchmark for tracking progress towards AGI. 2025 is going to be wild.
RESEARCH
-

ML Model-Serving Patterns and Best Practices Production Guide
By
–
#MachineLearning Model-Serving Patterns & Best Practices — Definitive guide to deploying, monitoring, & providing accessibility to #ML models in production: http://
amzn.to/3vNwKoT v/ @PacktPublishing ——
#AI #DeepLearning #DataScience #DataScientist #MLOps #CDO #DataEngineering -

Deep Reinforcement Learning Hands-On Guide: Q-Learning to RLHF
By
–
Deep #ReinforcementLearning Hands-On — A practical and easy-to-follow guide to RL from Q-learning and DQNs to PPO and RLHF: http://
amzn.to/3MV9o60 [Third Edition] v/ @PacktPublishing ———
#AI #MachineLearning #DeepLearning #DataScience #DataScientist
———
𝓚𝓮𝔂 𝓕𝓮𝓪𝓽𝓾𝓻𝓮𝓼: -
AGI Milestone: Model Fixes OpenAI Naming Issues
By
–
We’ll know we’ve reached AGI when the model can fix the OpenAI naming issues.
-
OpenAI Ships Major Updates Behind Scenes Work
By
–
so proud of the openai team for all that was just shipped. huge amount of work happening behind the scenes for each one of these launches.
-
o3 Model Architecture: Matrix Multiplications and Gradient Descent Training
By
–
o3 is a lot of matmuls trained with gradient descent.
-

Diffusion Models and Score-Based Generative Models Explained
By
–
Diffusion Models From Scratch | Score-Based Generative Models Explained | Math Explained https://
bit.ly/4gHxVs3 -
Computer Expected to Solve P vs NP or Riemann Hypothesis in 5 Years
By
–
I expect P vs NP or Riemann Hypothesis will be solved by a computer in the next 5 years, which is also when I will declare AGI.
-

Comprehensive Machine Learning Python Book for Algorithmic Trading
By
–
Comprehensive 822-page book on #MachineLearning with #Python for #AlgorithmicTrading (useful in many other business & industry #TimeSeries applications): http://
amzn.to/3aCyVwY by @ml4trading ——
#IoT #IIoT #Edge #AI #DataScience #EdgeComputing #Industry40 #EdgeAI #Forecasting -
AI Project Accused of Misleading Speedup Claims and Benchmarks
By
–
oh boy, it gets worse, looks like their speedup claims are very sleazy. do approximations for themselves and accurate sim for their competition to show themselves better: https://
x.com/stone_tao/stat
us/1870243004730225009
…
This project should go into "hustle bros", unfortunately. Not sure why a decent
