Deep learning
A branch of machine learning that uses neural networks with many layers to learn features directly from raw signals such as images, audio or vibration.
Deep learning uses multi-layer neural networks that learn their own feature representations instead of relying on hand-engineered inputs. This is why it dominates computer vision, speech and complex time-series work: given enough labeled data, the network discovers which patterns matter. In IoT it powers camera-based inspection, sound-based fault detection and gesture recognition.
The price is compute and data. Training large networks takes GPUs and large datasets, and inference on an edge device often needs an accelerator plus model compression. For a simple threshold or a linear sensor relationship, classical statistics are cheaper and easier to explain. Our guide Edge AI explained discusses when a deep model earns its place on a device.
Part of the IoT Glossary. Spotted an error or a missing term? Tell us.