About to try and train a neural network on a smol, binarized version of MNIST....wish me luck 😅 place your bets on train/test accuracy! I reckon at least 10% ;) #machinelearning
Yesterday, Amazon released a new open-source project, Chronos - a family of pre-trained time series forecasting models based on language model architectures.
We really need a different term than ‘AI’ to describe private, on-device machine learning that benefits individuals but (a) doesn’t violate anyone’s privacy, (b) doesn’t destroy the environment and, (c) doesn’t enrich smug Silicon Valley tech douchebros like Sam Altman.
Fabric is a new open-source project that provides a framework to support AI applications. The goal of Fabric is to unify the communication with AI agents (e.g., LLMs, etc.) by creating a library of Patterns (e.g., prompts) for day-to-day use cases.
This led to the publication of the following report by the @w3c team yesterday:
"AI & the Web: Understanding and managing the impact of Machine Learning models on the Web" https://www.w3.org/reports/ai-web-impact/
The MLflow for Machine Learning Development course by Manuel Gil provides a great introduction to the MLflow Python library 🐍. The course focuses on the MLflow core functionality and workflow and covers the following topics:
✅ Setting MLflow
✅ Creating and working with experiences
✅ Logging metadata (parameters, score, etc.)
✅ Model registry
✅ Model tuning
✅ MLflow project demo
Allez, petit article qui va bien, tapé à l'arrache, mais qui peut vous intéresser. Comment j'ai utilisé une #IA, locale, pour générer de la data fictive.
Code fourni en bas de l'article. Et n'hésitez pas à réagir dans la section commentaire !
No wonder #Amazon was trying to get me to give them more reviews.
An Amazon #chatbot that's supposed to surface useful information from customer reviews will also recommend a variety of racist books, lie about working conditions at Amazon, more. They just put a full, easy to abuse chatbot on their site
(1/2) Foundation Models & Generative AI Course - MIT Course 🚀
The course by Rickard Brüel Gabrielsson is a crash course on foundation models. This is the second version of the course, and it covers topics such as:
✅ Introduction to foundation models
✅ Different algorithms (ChatGPT, Stable-Diffusion & Dall-E)
✅ Supervised learning
✅ Neural networks
✅ Reinforcement learning
✅ Self-supervised learning
✅ Auto-encoders
One use of LLMs that I haven't seen mentioned before is to use them as a sounding board for your own ideas.
By discussing your concept with an LLM, you can gain fresh perspectives through its generated responses.
In this context, the LLM's actual comprehension is irrelevant. The purpose lies in its ability to spark new thought processes by prompting you with unexpected framings or questions.
Definitely recommend trying this trick next time you're writing something.
The ML for Beginners is a course by Microsoft that covers the foundation of machine learning. As the name implies, this 12-week course is for beginners and provides intro to the following topics:
✅ Fairness and machine learning
✅ Regression and classification
✅ Clustering
✅ Natural language processing
✅ Time series forecasting
✅ Reinforcement learning
sounds like the much heralded job of the future, "prompt engineer" is no longer needed 😅
"Battle and his collaborators found that in almost every case, this automatically [AI generated] generated prompt did better than the best prompt found through trial-and-error. And, the process was much faster, a couple of hours rather than several days of searching."
(1/2) Data Analysis for Social Scientists - MIT Course 🚀
MIT released a new course - Data Analysis for Social Scientists (MIT 14.310x), focusing on data analysis for social fields of science such as economics, public policy, and culture. This ten weeks course, by Prof. Esther Duflo and Dr. Sara Ellison, covers topics such as:
(1/2) Apple open source a new Python library for simulation framework for accelerating research in Private Federated Learning.
The library - pfl is a Python framework developed at Apple to empower researchers to run efficient simulations with privacy-preserving federated learning (FL) and disseminate the results of their research in FL.