MultiplicativeLSTM

MultiplicativeLSTM(input_size => hidden_size;
    return_state=false,
    kwargs...)

Multiplicative long short term memory network ^[Krause2017]. See MultiplicativeLSTMCell for a layer that processes a single sequence.

Arguments

• input_size => hidden_size: input and inner dimension of the layer.

Keyword arguments

• init_kernel: initializer for the input to hidden weights. Default is glorot_uniform.
• init_recurrent_kernel: initializer for the hidden to hidden weights. Default is glorot_uniform.
• init_multiplicative_kernel: initializer for the multiplicative to hidden weights. Default is glorot_uniform.
• bias: include a bias or not. Default is true.
• return_state: Option to return the last state together with the output. Default is false.

Equations

\[\begin{aligned} \mathbf{m}(t) &= \left( \mathbf{W}^{m}_{ih} \mathbf{x}(t) \right) \circ \left( \mathbf{W}^{m}_{hh} \mathbf{h}(t-1) \right), \\ \hat{\mathbf{h}}(t) &= \mathbf{W}^{h}_{ih} \mathbf{x}(t) + \mathbf{W}^{h}_{mh} \mathbf{m}(t) + \mathbf{b}^{h}, \\ \mathbf{i}(t) &= \sigma\left( \mathbf{W}^{i}_{ih} \mathbf{x}(t) + \mathbf{W}^{i}_{mh} \mathbf{m}(t) + \mathbf{b}^{i} \right), \\ \mathbf{o}(t) &= \sigma\left( \mathbf{W}^{o}_{ih} \mathbf{x}(t) + \mathbf{W}^{o}_{mh} \mathbf{m}(t) + \mathbf{b}^{o} \right), \\ \mathbf{f}(t) &= \sigma\left( \mathbf{W}^{f}_{ih} \mathbf{x}(t) + \mathbf{W}^{f}_{mh} \mathbf{m}(t) + \mathbf{b}^{f} \right), \\ \mathbf{c}(t) &= \mathbf{f}(t) \circ \mathbf{c}(t-1) + \mathbf{i}(t) \circ \tanh\left( \hat{\mathbf{h}}(t) \right), \\ \mathbf{h}(t) &= \tanh\left( \mathbf{c}(t) \right) \circ \mathbf{o}(t) \end{aligned}\]

Forward

multiplicativelstm(inp, (state, cstate))
multiplicativelstm(inp)

Arguments

• inp: The input to the multiplicativelstm. It should be a vector of size input_size x len or a matrix of size input_size x len x batch_size.
• (state, cstate): A tuple containing the hidden and cell states of the MultiplicativeLSTM. They should be vectors of size hidden_size or matrices of size hidden_size x batch_size. If not provided, they are assumed to be vectors of zeros, initialized by Flux.initialstates.

Returns

• New hidden states new_states as an array of size hidden_size x len x batch_size. When return_state = true it returns a tuple of the hidden stats new_states and the last state of the iteration.