MiRU2

MiRU2(input_size => hidden_size, [activation];
    init_kernel = glorot_uniform,
    init_recurrent_kernel = glorot_uniform,
    bias = true, recurrent_bias = true,
    independent_recurrence = false, integration_mode = :addition,
    update_coefficient = 0.5, reset_coefficient = 0.5)

Minion gated recurrent unit 2 (Zyarah and Kudithipudi, 2026). See MiRU2 for a layer that processes entire sequences.

Arguments

• input_size => hidden_size: input and inner dimension of the layer.
• activation: activation function. Default tanh_fast.

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.
• bias: include input to recurrent bias or not. Default is true.
• recurrent_bias: include recurrent 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.
• update_coefficient: controls the dynamic update of the hidden states. Default is 0.5.
• reset_coefficient: determines how much of the previous hidden state should be forgotten or reset before combining it with the new input. Default is 0.5.

Equations

\[\begin{aligned} \tilde{\mathbf{h}}(t) &= \tanh\!\left( \mathbf{W}_{h}\mathbf{x}(t) + \mathbf{b}_{h} + \mathbf{U}_{h}\!\left(\boldsymbol{\theta} \odot \mathbf{h}(t-1)\right) \right), \\ \mathbf{h}(t) &= \boldsymbol{\lambda} \odot \mathbf{h}(t-1) + \left(1 - \boldsymbol{\lambda}\right) \odot \tilde{\mathbf{h}}(t) \end{aligned}\]

Forward

miru(inp, (state, cstate))
miru(inp)

Arguments

• inp: The input to the miru. 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 MiRU2. 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.