A collection of the the best ML and AI news every week (research, news, resources)

This Repository contains a curated weekly collection of AI & ML news, refined into clear, readable insights and compiled into a clean, shareable PDF. It focuses on cutting through noise, highlighting what truly matters each week in the AI space. It serves as a structured weekly exercise to understand and summarize AI developments.

A collection of the the best ML and AI news every week (research, news, resources)

Week 1 project at the ml.institute programme. Building a tool to predict the upvote score of posts on the Hacker News website https://news.ycombinator.com/ using just their titles.

A service that fetches updates to ESPN, Yahoo, and Fox Sports NFL News RSS Feeds and also grabs the current NFL week's batch of scores as well as the lines and odds for ML, ATS, and O/U bets

Problem Statement: We have a list of trades of different simulations (eash is numbered and based on different settings). Each simulation represents a different permutation of settings used to run them. These are results of a trading algorithm which decides when and how to buy or sell different assets. We want to know what permutation will work for Wednesday 15:00 (for example) and we assume that each day of the week and hour of the day has a different profile as certain activities are based on news which are the same hours of the days. etc. Selecting the best permutation (best by aggregated profit & loss) for a future period is needed by selecting day of the week and time of the day. You can take any feature you want, and you can use anything which will make the prediction better (you can even download Gold prices and use it's value and volatility if you think this might help) so long as your prediction is the profit/loss. The different ML method / model is upon your selection.