Sports Analytics Weekly by kubeia.io - 1/2023
- 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
💰 Quantitative Finance
How to Build an Exchange
janestreet.com
This talk looks at the question of how to design an exchange through the lens of JX, a crossing engine we built at Jane Street in the last two years. Performance plays an interesting role in this design, in that, although the end-to-end latency of the system is not important in and of itself, the ability of individual components of JX to handle messages rates in the 500k/sec range with latencies in the single-digit microseconds helped us build a replicated system that is both simple and robust.
When Point Forecasts Are Completely Useless
sarem-seitz.com - Sarem Seitz
While point forecasts are very popular, be aware of some unlucky pitfalls.
🤖 Machine Learning
Welcome to the 🤗 Deep Reinforcement Learning Course
huggingface.co
This course will teach you about Deep Reinforcement Learning from beginner to expert. It’s completely free and open-source!
A Big Year For AI
sebastianraschka.com - Sebastian Raschka
To mark the start of the new year, this month's issue will feature a review of the top ten papers I've read in 2022.
Machine Learning & AI Open Source Highlights 2022
sebastianraschka.com - Sebastian Raschka
Recently, I shared the top 10 papers that I read in 2022. As a follow-up, I am compiling a list of my favorite 10 open-source releases that I discovered, used, or contributed to in 2022.
Machine Learning and The Cross-Section of Emerging Market Stock Returns
ssrn.com
This paper compares various machine learning models to predict the cross-section of emerging market stock returns. We document that allowing for non-linearities and interactions leads to economically and statistically superior out-of-sample returns compared to traditional linear models. Although we find that both linear and machine learning models show higher predictability for stocks associated with higher limits to arbitrage, we also show that this effect is less pronounced for non-linear models. Furthermore, significant net returns can be achieved when accounting for transaction costs, short-selling constraints, and limiting our investment universe to big stocks only.
The 2022 AI Recap - A Review
louisbouchard.ai
A curated list of the latest breakthroughs in AI by release date with a clear video explanation, link to a more in-depth article, and code.
What is the Kernel Trick?
eranraviv.com
The kernel trick is helpful for expanding any data from its original dimension to higher dimension. Fine, who cares?. Well, in life we often benefit from dimension expansion. Higher dimension makes for a richer data representation, which statistical models can exploit to create better predictions.
🚀 Engineering
How to Optimize a CUDA Matmul Kernel for cuBLAS-like Performance: a Worklog
siboehm.com
In this post, I’ll iteratively optimize an implementation of matrix multiplication written in CUDA. My goal is not to build a cuBLAS replacement, but to deeply understand the most important performance characteristics of the GPUs that are used for modern deep learning. This includes coalescing global memory accesses, shared memory caching and occupancy optimizations, among others.