Tune My Adam, Please!
A high level overview of our paper ‘Tune My Adam, Please!’, where we introduce a surrogate model for HPO specialized for the Adam optimizer.
Recent posts
Is there such thing as too much HPO?
A discussion of the paper: ‘Overtuning in Hyperparameter Optimization’ by L. Schneider, B. Bischl, and M. Feurer.
Code Walkthrough - RAG with LangChain
A practical walkthrough of RAG with LangChain: build a vector database, wire a retriever into a prompt, and serve a Streamlit PDF Q&A app.
Deconstructing Dremel: Google’s Tool For Interactive Analysis of Web-Scale Datasets
An in-depth look at Dremel, Google’s interactive analysis system for web-scale datasets.
The ML Validation Playbook: A Primer of Common Validation Techniques
A primer of common validations techniques before we discuss the ‘Overtuning in Hyperparameter Optimization’ paper by L. Schneider et al.