ML on the client: training on squirrels
summary
Covers using machine learning and deep learning models directly in client-side applications: explains how neural networks work and then shows practical deployment options (where and how to run models in the browser or on-device). Discusses when client-side ML is beneficial and when it’s better to keep inference on the server.
AI-written summary · about
chapters (20)
- 0:00Welcome
- 1:05Introduction
- 3:13ML matters for frontend
- 5:42Declarative vs imperative ML
- 8:32Neural networks work
- 11:03Tensors
- 12:31Image convolution layers
- 15:21Data representation
- 18:13Training process steps
- 19:16Squirrel vs cat classifier
- 22:07Model layer analysis
- 25:18Model size vs accuracy
- 27:08ML in browser
- 29:18Model performance
- 31:25WebGPU and WebGL acceleration
- 35:40Browser ML solutions
- 36:24Q&A
- 38:53Full client-side ML?
- 41:19When to use client ML
- 44:11Conclusion
related talks
19mML at Client Side
23mNode.js = Machine Learning
26mMachine Learning and AI for Angular Developers
34mA Powerful Introduction to Machine Learning with Js
15mThe future of machine learning using JavaScript
34mA Powerful Introduction to Machine Learning with Js
37mMachine Learning Principles, with JavaScript?
41mPractical Machine Learning for Front-End Developers