Figure 5.

Architecture of a LSTM network for sensor-based HAR (left). The intermediate and last LSTM layers are organized in a many-to-many and many-to-one layout, respectively. Each LSTM layer is composed of LSTM cells (right). ${\mathit{x}}_t$, ${\mathit{m}}_t$ and ${\mathit{c}}_t$ represent the cell input, memory, and output at time t, respectively. In addition, ⨁ and ⨂ refer to the element-wise addition and multiplication, respectively. ${\mathit{W}}_{*}$, ${\mathit{U}}_{*}$, ${\mathit{b}}_{*}$ and ${\sigma }_{*}$ designate internal weight matrices, bias vector and activation function (for gate $*\in \{i,o,f\}$). ${\sigma }_1$ and ${\sigma }_2$ are internal activation functions applied on the input and memory of the cell, respectively.