Part IV: Deep Learning for Sensor Time Series
Chapter 13  [C]

Neural Representations for Sensor Streams

Placeholder chapter page. Sections are produced by the book-skills pipeline.

Sections

  1. 13.1 Sensor windows as tensors; channel and axis structure
  2. 13.2 1D convolutions and temporal pooling
  3. 13.3 Multiresolution and dilated features
  4. 13.4 Learned vs fixed filterbanks
  5. 13.5 Normalization and augmentation for sensors
  6. 13.6 Inductive biases for time series
  7. 13.7 Debugging and inspecting learned filters

Lab 13

train a 1D CNN for human activity recognition; visualize learned filters.