A Discussion of ‘Adversarial Examples Are Not Bugs, They Are Features’: Robust Feature Leakage
Ilyas et al. report a surprising result: a model trained on adversarial examples is effective…
Easy Agentic Tool Calling with Gemma 4
# Introduction In a recent article on Machine Learning Mastery, we built a tool-calling agent…
Hybrid AI: Combining Deterministic Analytics with LLM Reasoning
Introduction an agentic AI network for my company that advises manufacturing plants on how to…
Two Examples of Useful, Non-Robust Features
A Discussion of 'Adversarial Examples Are Not Bugs, They Are Features': Two Examples of Useful,…
100 things we announced at Google I/O 2026
AI Inbox68. Earlier this year, we introduced AI Inbox as a new view in Gmail…
How to Get the Most Out of Claude Cowork
# Introduction to Claude Cowork Most people who use Claude like a smarter search engine.…
Six Choices Every AI Engineer Has to Make (and Nobody Teaches)
teach you how to make a model accurate. They rarely teach you the decisions that…
Adversarially Robust Neural Style Transfer
A figure in Ilyas, et. al. that struck me as particularly interesting was the following…
TurboQuant: Is the Compression and Performance Worth the Hype?
# Introduction TurboQuant is a novel algorithmic suite and library recently launched by Google. Its…