Stephen Holmes’ Post

In retrospect, this interview with Geordie Rose I recorded in 2013, has turned out to be prophetic: Geordie Rose: Machine Learning is Progressing Faster Than You Think https://lnkd.in/dkTVTwB

Better Algorithms and QC-ML are going to help us solve more complex use cases and take AI to the next level. Particularly, CQ Algorithms/Quantum for machine learning https://lnkd.in/gNZ9UeDn

Delighted to announce my first machine learning paper! You may have seen the writing on the wall when I published my ML/compiler blog posts, but now I'm all in! In this paper we manage to quantize BitNet even further, producing very small and very fast models: https://lnkd.in/e3pvGSyi #machinelearning #ml #llm #bitnet

An *amazing* article by Maxime Wolf about the curse of dimensionality in machine learning models, and how the solution is linked to a very intriguing property of spheres in higher dimensions: https://lnkd.in/gHTCD9FY

https://lnkd.in/dH25SPUQ A new approach, called algorithms with predictions, uses the power of machine learning to improve algorithms.  #algorithms #algorithm #algorithmicthinking #algorithmdesign #machinelearning #machinelearningalgorithms

With the rise of LLMs and CNNs building deeper models is an important part of solving the matching learning puzzle. What if there is a way to (1) simultaneously reduce the parameters in your model AND (2) increase the model depth? 1×1 convolutions are your answer. These help you build: - faster (training and inference time), - leaner (smaller memory footprint, smaller GPU to run them), and - smarter machine learning models (equally performant if not better). Learn what are 1×1 convolutions and when to use them. https://lnkd.in/dtzaXePs

The origins and applications of machine learning

Hello Connections🎓 This is my certificate from edx upon the completion of Machine Learning course.This course has covered all the algorithms,machine learning models and applications of ML.

Hello people! Introduction to Machine Learning: Gained foundational knowledge in machine learning concepts, algorithms, and techniques.

I republished: A Litter Model using Active Inference Using Bayesian Inference and RxInfer to estimate daily litter events https://lnkd.in/geCEvNRN I experimented with the beginnings of a 'Unified Framework' for bayesian machine learning. I like the economy of expression that comes from that, as well as the associated standardization. It does come with some 'clumsiness' sometimes, but to me, it is worth it. This potential framework, for now referred to as KUF, to keep it apart from PUF, is an overkill for the problem in this notebook. However, it serves as a template for more substantial projects in the future. The KUF is informed by the CRISP-DM methodology, the Powell Unified Framework (PUF), formerly at Princeton, the work of professor Bert de Vries at Eindhoven, as well as the Active Inference literature that I have used so far. Your comments are welcome.