Sports Analytics Weekly by kubeia.io - 33/2023
- Week 52
- Week 51
- Week 49
- Week 48
- Week 47
- Week 46
- Week 45
- Week 44
- Week 43
- Week 42
- Week 41
- Week 40
- Week 39
- Week 38
- Week 37
- Week 36
- Week 35
- Week 34
- Week 33
- Week 32
- Week 31
- Week 30
- Week 29
- Week 28
- Week 27
- Week 26
- Week 25
- Week 24
- Week 23
- Week 22
- Week 21
- Week 20
- Week 19
- Week 18
- Week 16
- Week 14
- Week 13
- Week 12
- Week 10
- Week 9
- Week 6
- Week 5
- Week 4
- Week 3
- Week 1
š Sports Analytics
Fantasy Premier League: Explaining āthe templateā ā and why you should back it
theathletic.com - Holly Shand
In serious Fantasy Premier League circles, āthe templateā is regularly discussed ā and is often seen as a dirty phrase.That leaves many managers outside of those groups wondering what the template is and whether itās a good thing or not.
š¤ Machine Learning
The Mathematics of Training LLMs ā with Quentin Anthony of Eleuther AI
latent.space
Breaking down the viral Transformers Math 101 article and high performance distributed training for Transformers-based architectures (or "How I Learned to Stop Handwaving and Make the GPU go brrrrrr")
Probabilistic Machine Learning: Advanced Topics
github.io - Kevin P. Murphy
This book is a sequel to [Mur22]. and provides a deeper dive into various topics in machine learning (ML).
AutoGluon-TimeSeries: AutoML for Probabilistic Time Series Forecasting
arxiv.org - Oleksandr Shchur, Caner Turkmen, Nick Erickson, Huibin Shen, Alexander Shirkov, Tony Hu, Yuyang Wang
We introduce AutoGluon-TimeSeries - an open-source AutoML library for probabilistic time series forecasting. Focused on ease of use and robustness, AutoGluon-TimeSeries enables users to generate accurate point and quantile forecasts with just 3 lines of Python code. Built on the design philosophy of AutoGluon, AutoGluon-TimeSeries leverages ensembles of diverse forecasting models to deliver high accuracy within a short training time. AutoGluon-TimeSeries combines both conventional statistical models, machine-learning based forecasting approaches, and ensembling techniques. In our evaluation on 29 benchmark datasets, AutoGluon-TimeSeries demonstrates strong empirical performance, outperforming a range of forecasting methods in terms of both point and quantile forecast accuracy, and often even improving upon the best-in-hindsight combination of prior methods.