Machine Learning
Explainers and news on machine learning: how models are trained, deep learning, synthetic data, NLP, open-source tools and running AI locally.
Stories from Machine Learning come first, newest first, then stories from connected topics (Deep Learning, Guides, Generative AI, AI, AI, AI for science), each labelled with where it comes from. This view is not indexed by search engines.
Linköping Team Uses AI to Find 33 Overlooked MXenes, Makes 11 New
A Linköping-led team used machine learning to spot 33 overlooked MXenes in old databases, then made 11 new rare-earth MXenes in minutes. Lab-stage work.
Latest
Mila and NRCan AI Maps Wildfire Hazard Hundreds of Times Faster
Mila and Natural Resources Canada say an AI copy of the BurnP3+ simulator maps wildfire hazard in seconds. The preprint is not yet peer reviewed.
AI-Guided Lab Finds Hydrogen Catalysts It Calls Comparable to Platinum
A UCLA-led robotic lab guided by interpretable AI tested about 1,200 recipes and found earth-abundant hydrogen catalysts it calls comparable to Pt.
Quick Guide: 7 Steps to Run Trading-Bot Statistics on a Grok Bot
Seven steps to run GARCH, regime and walk-forward models on a Grok Bot's cloud computer, from a virtual environment to backtests, keys and backups.
A Grok Bot Can Run Statistical Models on Its Own Cloud Computer
A Grok Bot's cloud computer can install Python stats libraries and fit models in seconds, but has no GPU, no published limits and shared keys.
insitro and AstraZeneca Sign AI Collaboration to Hunt Drugs for MASH and Cardiomyopathies
insitro gets $40m upfront and is eligible for up to about $1.46bn in potential milestones in an AstraZeneca AI deal for MASH and heart disease.
AWS Makes Its AI/ML Skill a Core Part of the Agent Toolkit, Taking Coding Agents From Inference Tuning to Full SageMaker Workflows
AWS's AI/ML skill is now a core Agent Toolkit skill, so coding agents can select, train, optimise and deploy SageMaker AI models in plain language.
Meta Says It Turns Idle AI Inference Capacity Into Better Ads Recommendations
Meta's engineers describe a shared opportunistic capacity pool that uses idle inference capacity to rank more ads and serve stronger models.
Google Open-Sources ML Drift, the GPU Engine Inside LiteRT, Under Apache 2.0
Google has open-sourced ML Drift, the GPU engine inside LiteRT, under Apache 2.0. The 40%, 30% and 12% figures are Google's own claims; the old TFLite GPU delegate gets no new features.
Timer-M1 Preprint Says It Ranks First on Two Forecasting Benchmarks, With Its Authors’ Own Caveats
A preprint says Timer-M1, a forecasting model, ranks first on FEV and TIME. Its authors call the TIME margin over TimesFM-3 narrow. Not peer reviewed.
Synthetic Data: The Complete Guide to AI’s Secret Weapon in 2026
The $285 Billion Secret: How Fake Data Is Powering Real AIDate: April 23, 2026
Category: AI / Technology
Reading Time: 10 minutes The Invisible...
Latest on TSN
Explore the TSN Atlas: Every topic we cover, mapped. Click a bubble to dive in.