This layer creates a convolution kernel that is convolved
(actually cross-correlated) with the layer input to produce a tensor of
outputs. It may also include a bias addition and activation function
on the outputs. It assumes the kernel and/or bias are drawn from distributions.
layer_conv_1d_flipout(
object,
filters,
kernel_size,
strides = 1,
padding = "valid",
data_format = "channels_last",
dilation_rate = 1,
activation = NULL,
activity_regularizer = NULL,
trainable = TRUE,
kernel_posterior_fn = tfp$layers$util$default_mean_field_normal_fn(),
kernel_posterior_tensor_fn = function(d) d %>% tfd_sample(),
kernel_prior_fn = tfp$layers$util$default_multivariate_normal_fn,
kernel_divergence_fn = function(q, p, ignore) tfd_kl_divergence(q, p),
bias_posterior_fn = tfp$layers$util$default_mean_field_normal_fn(is_singular = TRUE),
bias_posterior_tensor_fn = function(d) d %>% tfd_sample(),
bias_prior_fn = NULL,
bias_divergence_fn = function(q, p, ignore) tfd_kl_divergence(q, p),
...
)a Keras layer
What to compose the new Layer instance with. Typically a
Sequential model or a Tensor (e.g., as returned by layer_input()).
The return value depends on object. If object is:
missing or NULL, the Layer instance is returned.
a Sequential model, the model with an additional layer is returned.
a Tensor, the output tensor from layer_instance(object) is returned.
Integer, the dimensionality of the output space (i.e. the number of filters in the convolution).
An integer or list of a single integer, specifying the length of the 1D convolution window.
An integer or list of a single integer,
specifying the stride length of the convolution.
Specifying any stride value != 1 is incompatible with specifying
any dilation_rate value != 1.
One of "valid" or "same" (case-insensitive).
A string, one of channels_last (default) or
channels_first. The ordering of the dimensions in the inputs.
channels_last corresponds to inputs with shape (batch, length, channels) while channels_first corresponds to inputs with shape
(batch, channels, length).
An integer or tuple/list of a single integer, specifying
the dilation rate to use for dilated convolution.
Currently, specifying any dilation_rate value != 1 is
incompatible with specifying any strides value != 1.
Activation function. Set it to None to maintain a linear activation.
Regularizer function for the output.
Whether the layer weights will be updated during training.
Function which creates tfd$Distribution instance representing the surrogate
posterior of the kernel parameter. Default value: default_mean_field_normal_fn().
Function which takes a tfd$Distribution instance and returns a representative
value. Default value: function(d) d %>% tfd_sample().
Function which creates tfd$Distribution instance. See default_mean_field_normal_fn docstring for required
parameter signature. Default value: tfd_normal(loc = 0, scale = 1).
Function which takes the surrogate posterior distribution, prior distribution and random variate
sample(s) from the surrogate posterior and computes or approximates the KL divergence. The
distributions are tfd$Distribution-like instances and the sample is a Tensor.
Function which creates a tfd$Distribution instance representing the surrogate
posterior of the bias parameter. Default value: default_mean_field_normal_fn(is_singular = TRUE) (which creates an
instance of tfd_deterministic).
Function which takes a tfd$Distribution instance and returns a representative
value. Default value: function(d) d %>% tfd_sample().
Function which creates tfd instance. See default_mean_field_normal_fn docstring for required parameter
signature. Default value: NULL (no prior, no variational inference)
Function which takes the surrogate posterior distribution, prior distribution and random variate sample(s)
from the surrogate posterior and computes or approximates the KL divergence. The
distributions are tfd$Distribution-like instances and the sample is a Tensor.
Additional keyword arguments passed to the keras::layer_dense constructed by this layer.
This layer implements the Bayesian variational inference analogue to
a dense layer by assuming the kernel and/or the bias are drawn
from distributions.
By default, the layer implements a stochastic forward pass via sampling from the kernel and bias posteriors,
outputs = f(inputs; kernel, bias), kernel, bias ~ posterior
where f denotes the layer's calculation. It uses the Flipout
estimator (Wen et al., 2018), which performs a Monte Carlo approximation
of the distribution integrating over the kernel and bias. Flipout uses
roughly twice as many floating point operations as the reparameterization
estimator but has the advantage of significantly lower variance.
The arguments permit separate specification of the surrogate posterior
(q(W|x)), prior (p(W)), and divergence for both the kernel and bias
distributions.
Upon being built, this layer adds losses (accessible via the losses
property) representing the divergences of kernel and/or bias surrogate
posteriors and their respective priors. When doing minibatch stochastic
optimization, make sure to scale this loss such that it is applied just once
per epoch (e.g. if kl is the sum of losses for each element of the batch,
you should pass kl / num_examples_per_epoch to your optimizer).
You can access the kernel and/or bias posterior and prior distributions
after the layer is built via the kernel_posterior, kernel_prior,
bias_posterior and bias_prior properties.
Other layers:
layer_autoregressive(),
layer_conv_1d_reparameterization(),
layer_conv_2d_flipout(),
layer_conv_2d_reparameterization(),
layer_conv_3d_flipout(),
layer_conv_3d_reparameterization(),
layer_dense_flipout(),
layer_dense_local_reparameterization(),
layer_dense_reparameterization(),
layer_dense_variational(),
layer_variable()