Topic

On-Device Machine Learning

Running trained machine-learning models directly on end-user hardware such as phones, wearables and embedded boards, without sending inputs to a remote inference service.

Current clusters

build1 publisher

An unquantized model on the main thread cost a health app its screening feature

A dev.to walkthrough of on-device ML on Android traces one withdrawn screening feature back to main-thread inference and a float32 model, and prices what quantization and a calibration set would have saved.

Publishers:dev.to

Reality

Evidence34
Adoption20
Hype gap+28
Incentives55
Confidence45