incerto.bayesian.VariationalBayesNN

class incerto.bayesian.VariationalBayesNN(in_features, hidden_sizes, out_features, prior_std=1.0, num_samples=20)

Bases: BaseBayesianMethod

Variational Bayesian Neural Network (Bayes by Backprop).

Learns a distribution over weights using variational inference. Each weight has a learned mean and variance.

Note

variational_loss currently uses cross-entropy and therefore only supports classification tasks. For regression, provide a custom training loop with an appropriate likelihood.

Reference:

Blundell et al., “Weight Uncertainty in Neural Networks” (ICML 2015)

Parameters:

• in_features (int) – Input dimension

• hidden_sizes (List[int]) – List of hidden layer sizes

• out_features (int) – Output dimension

• prior_std (float) – Prior standard deviation (default: 1.0)

• num_samples (int) – Number of MC samples for prediction (default: 20)

Example

>>> model = VariationalBayesNN(784, [512, 256], 10)
>>> # Train with variational loss
>>> for x, y in train_loader:
...     loss = model.variational_loss(x, y, num_samples=10)
...     loss.backward()
>>> mean, variance = model.predict(x)

__init__(in_features, hidden_sizes, out_features, prior_std=1.0, num_samples=20)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

Parameters:

• in_features (int)

• hidden_sizes (List[int])

• out_features (int)

• prior_std (float)

• num_samples (int)

Methods

__init__(in_features, hidden_sizes, out_features)	Initialize internal Module state, shared by both nn.Module and ScriptModule.
add_module(name, module)	Add a child module to the current module.
apply(fn)	Apply fn recursively to every submodule (as returned by .children()) as well as self.
bfloat16()	Casts all floating point parameters and buffers to bfloat16 datatype.
buffers([recurse])	Return an iterator over module buffers.
children()	Return an iterator over immediate children modules.
compile(*args, **kwargs)	Compile this Module's forward using torch.compile().
cpu()	Move all model parameters and buffers to the CPU.
cuda([device])	Move all model parameters and buffers to the GPU.
double()	Casts all floating point parameters and buffers to double datatype.
eval()	Set the module in evaluation mode.
extra_repr()	Return the extra representation of the module.
float()	Casts all floating point parameters and buffers to float datatype.
forward(x)	Forward pass with sampled weights.
get_buffer(target)	Return the buffer given by target if it exists, otherwise throw an error.
get_extra_state()	Return any extra state to include in the module's state_dict.
get_parameter(target)	Return the parameter given by target if it exists, otherwise throw an error.
get_submodule(target)	Return the submodule given by target if it exists, otherwise throw an error.
half()	Casts all floating point parameters and buffers to half datatype.
ipu([device])	Move all model parameters and buffers to the IPU.
kl_divergence()	Compute KL divergence between posterior and prior.
load_state_dict(state_dict[, strict, assign])	Copy parameters and buffers from state_dict into this module and its descendants.
modules()	Return an iterator over all modules in the network.
mtia([device])	Move all model parameters and buffers to the MTIA.
named_buffers([prefix, recurse, ...])	Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.
named_children()	Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.
named_modules([memo, prefix, remove_duplicate])	Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.
named_parameters([prefix, recurse, ...])	Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.
parameters([recurse])	Return an iterator over module parameters.
predict(x[, return_samples, normalize_output])	Variational prediction with uncertainty.
register_backward_hook(hook)	Register a backward hook on the module.
register_buffer(name, tensor[, persistent])	Add a buffer to the module.
register_forward_hook(hook, *[, prepend, ...])	Register a forward hook on the module.
register_forward_pre_hook(hook, *[, ...])	Register a forward pre-hook on the module.
register_full_backward_hook(hook[, prepend])	Register a backward hook on the module.
register_full_backward_pre_hook(hook[, prepend])	Register a backward pre-hook on the module.
register_load_state_dict_post_hook(hook)	Register a post-hook to be run after module's load_state_dict() is called.
register_load_state_dict_pre_hook(hook)	Register a pre-hook to be run before module's load_state_dict() is called.
register_module(name, module)	Alias for add_module().
register_parameter(name, param)	Add a parameter to the module.
register_state_dict_post_hook(hook)	Register a post-hook for the state_dict() method.
register_state_dict_pre_hook(hook)	Register a pre-hook for the state_dict() method.
requires_grad_([requires_grad])	Change if autograd should record operations on parameters in this module.
set_extra_state(state)	Set extra state contained in the loaded state_dict.
set_submodule(target, module[, strict])	Set the submodule given by target if it exists, otherwise throw an error.
share_memory()	See torch.Tensor.share_memory_().
state_dict(*args[, destination, prefix, ...])	Return a dictionary containing references to the whole state of the module.
to(*args, **kwargs)	Move and/or cast the parameters and buffers.
to_empty(*, device[, recurse])	Move the parameters and buffers to the specified device without copying storage.
train([mode])	Set the module in training mode.
type(dst_type)	Casts all parameters and buffers to dst_type.
variational_loss(x, y[, num_samples, kl_weight])	Compute variational loss (ELBO).
xpu([device])	Move all model parameters and buffers to the XPU.
zero_grad([set_to_none])	Reset gradients of all model parameters.

Attributes

T_destination
call_super_init
dump_patches
training

__init__(in_features, hidden_sizes, out_features, prior_std=1.0, num_samples=20)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

Parameters:

• in_features (int)

• hidden_sizes (List[int])

• out_features (int)

• prior_std (float)

• num_samples (int)

forward(x)

Forward pass with sampled weights.

Parameters:

x (Tensor)

Return type:

Tensor

kl_divergence()

Compute KL divergence between posterior and prior.

Return type:

Tensor

variational_loss(x, y, num_samples=10, kl_weight=1.0)

Compute variational loss (ELBO).

Loss = E[NLL] + KL[q(w) || p(w)]

Parameters:

• x (Tensor) – Input tensor

• y (Tensor) – Target labels

• num_samples (int) – Number of samples for MC estimate

• kl_weight (float) – Weight for KL term

Return type:

Tensor

Returns:

Variational loss

predict(x, return_samples=False, normalize_output=True)

Variational prediction with uncertainty.

Parameters:

• x (Tensor) – Input tensor

• return_samples (bool) – If True, return all sampled predictions

• normalize_output (bool) – If True, apply softmax to 2-D multi-column outputs (i.e. treat them as logits). Set to False when the model already returns probabilities or when outputs are not classification logits.

Return type:

Union[Tuple[Tensor, Tensor], Tuple[Tensor, Tensor, Tensor]]

Returns:

Tuple of (mean_prediction, predictive_variance) If return_samples=True: (mean, variance, samples)