SCRNCell

SCRNCell(input_size => hidden_size;
    init_kernel = glorot_uniform,
    init_recurrent_kernel = glorot_uniform,
    init_context_kernel = glorot_uniform,
    bias = true, recurrent_bias = true, context_bias=true,
    independent_recurrence = false, integration_mode = :addition, alpha = 0.0)

Structurally contraint recurrent unit (Mikolov et al., 2014). See SCRN for a layer that processes entire sequences.

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_context_kernel: initializer for the context to hidden weights. Default is glorot_uniform.
• bias: include input to recurrent bias or not. Default is true.
• recurrent_bias: include recurrent to recurrent bias or not. Default is true.
• context_bias: include context to recurrent bias or not. Default is true.
• independent_recurrence: flag to toggle independent recurrence. If true, the recurrent to recurrent weights are a vector instead of a matrix. Default false.
• integration_mode: determines how the input and hidden projections are combined. The options are :addition and :multiplicative_integration. Defaults to :addition.
• alpha: structural contraint. Default is 0.0.

Equations

\[\begin{aligned} \mathbf{s}(t) &= (1 - \alpha) \, \mathbf{W}_{ih}^{s} \mathbf{x}(t) + \alpha \, \mathbf{s}(t-1) \\ \mathbf{h}(t) &= \sigma\left( \mathbf{W}_{ih}^{h} \mathbf{s}(t) + \mathbf{W}_{hh}^{h} \mathbf{h}(t-1) + \mathbf{b}^{h} \right) \\ \mathbf{y}(t) &= f\left( \mathbf{W}_{hh}^{y} \mathbf{h}(t) + \mathbf{W}_{ih}^{y} \mathbf{s}(t) + \mathbf{b}^{y} \right) \end{aligned}\]

Forward

scrncell(inp, (state, cstate))
scrncell(inp)

Arguments

• inp: The input to the scrncell. It should be a vector of size input_size or a matrix of size input_size x batch_size.
• (state, cstate): A tuple containing the hidden and cell states of the SCRNCell. 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

• A tuple (output, state), where output = new_state is the new hidden state and state = (new_state, new_cstate) is the new hidden and cell state. They are tensors of size hidden_size or hidden_size x batch_size.