This is a Pytorch implementation of Poisson Identifiable VAE (pi-VAE), used to construct latent variable models of neural activity while simultaneously modeling the relation between the latent and task variables (non-neural variables, e.g. sensory, motor, and other externally observable states).
The original implementation by Ding Zhou and Xue-Xin Wei in Tensorflow 1.13 is available here.
Another Pytorch implementation by Lyndon Duong is available here.
A special thank you to Zhongxuan Wu who helped in the design and testing of this implementation.
pip install pi-vae-pytorch
For those interested in modifying and testing the codebase, using an editable pip
installation is recommended:
# pi-vae-pytorch/
pip install -e .
pi-VAE is comprised of three main components: the encoder, the label prior estimator, and the decoder.
The Multi Layer Perceptron (MLP) is the primary building block of the aforementioned components. Each MLP used in this implementation is configurable by specifying the appropriate parameters when PiVAE
is initialized:
- number of hidden layers
- hidden layer dimension
- applied to all hidden layers within a given MLP
- hidden layer activation function
- applied to all non-output layer activations within a given MLP
The model's encoder is comprised of a single MLP, which learns the distribution q(z | x).
The model's label prior estimator learns the distribution p(z | u). In the discrete label regime this module is comprised of a single nn.Embedding
layer, while in the continuous label regime the module is comprised of a MLP.
The model's decoder learns to map a latent sample z
to the predicted firing rate of the latent sample. Inputs to the decoder are passed through the following submodules.
This module is comprised of a MLP which maps z
to the concatenation of z
and t(z)
.
Outputs from the NFlowLayer
are passed to a series of GINBlock
modules. Each GINBlock
is comprised of a specified number of AffineCouplingLayer
modules. Each AffineCouplingLayer
is comprised of a MLP and performs an affine coupling transformation.
import torch
from pi_vae_pytorch import PiVAE
model = PiVAE(
x_dim = 100,
u_dim = 3,
z_dim = 2,
discrete_labels=False
)
x = torch.randn(1, 100) # Size([n_samples, x_dim])
u = torch.randn(1, 3) # Size([n_samples, u_dim])
outputs = model(x, u) # Dict
-
x_dim
: int
Dimension of observationx
-
u_dim
: int
Dimension of labelu
-
z_dim
: int
Dimension of latentz
-
discrete_labels
: bool- Default:
True
Flag denoting
u
's label type -True
: discrete orFalse
: continuous. - Default:
-
encoder_n_hidden_layers
: int- Default:
2
Number of hidden layers in the MLP of the model's encoder.
- Default:
-
encoder_hidden_layer_dim
: int- Default:
120
Dimensionality of each hidden layer in the MLP of the model's encoder.
- Default:
-
encoder_hidden_layer_activation
: nn.Module- Default:
nn.Tanh
Activation function applied to the outputs of each hidden layer in the MLP of the model's encoder.
- Default:
-
decoder_n_gin_blocks
: int- Default:
2
Number of GIN blocks used within the model's decoder.
- Default:
-
decoder_gin_block_depth
: int- Default:
2
Number of AffineCouplingLayers which comprise each GIN block.
- Default:
-
decoder_affine_input_layer_slice_dim
: int- Default None (corresponds to
x_dim // 2
)
Index at which to split an n-dimensional input x.
- Default None (corresponds to
-
decoder_affine_n_hidden_layers
: int- Default:
2
Number of hidden layers in the MLP of the model's encoder.
- Default:
-
decoder_affine_hidden_layer_dim
: int- Default:
None
(corresponds tox_dim // 4
)
Dimensionality of each hidden layer in the MLP of each AffineCouplingLayer.
- Default:
-
decoder_affine_hidden_layer_activation
: nn.Module- Default:
nn.ReLU
Activation function applied to the outputs of each hidden layer in the MLP of each AffineCouplingLayer.
- Default:
-
decoder_nflow_n_hidden_layers
: int- Default:
2
Number of hidden layers in the MLP of the decoder's NFlowLayer.
- Default:
-
decoder_nflow_hidden_layer_dim
: int- Default:
None
(corresponds tox_dim // 4
)
Dimensionality of each hidden layer in the MLP of the decoder's NFlowLayer.
- Default:
-
decoder_nflow_hidden_layer_activation
: nn.Module- Default:
nn.ReLU
Activation function applied to the outputs of each hidden layer in the MLP of the decoder's NFlowLayer.
- Default:
-
decoder_observation_model
: str- Default:
poisson
- One of
gaussian
orpoisson
Observation model used by the model's decoder.
- Default:
-
decoder_fr_clamp_min
: float- Default:
1E-7
- Only applied when
decoder_observation_model="poisson"
Mininimum threshold used when clamping decoded firing rates.
- Default:
-
decoder_fr_clamp_max
: float- Default:
1E7
- Only applied when
decoder_observation_model="poisson"
Maximum threshold used when clamping decoded firing rates.
- Default:
-
z_prior_n_hidden_layers
: int- Default:
2
- Only applied when
discrete_labels=False
Number of hidden layers in the MLP of the label prior estimator module.
- Default:
-
z_prior_hidden_layer_dim
: int- Default:
20
- Only applied when
discrete_labels=False
Dimensionality of each hidden layer in the MLP of the label prior estimator module.
- Default:
-
z_prior_hidden_layer_activation
: nn.Module- Default:
nn.Tanh
- Only applied when
discrete_labels=False
Activation function applied to the outputs of each hidden layer in the MLP of the label prior estimator module.
- Default:
A Dict
with the following items.
-
firing_rate
: Tensor- Size([n_samples, x_dim])
Predicted firing rates of
z_sample
. -
lambda_mean
: Tensor- Size([n_samples, z_dim])
Mean for each sample using label prior p(z | u).
-
lambda_log_variance
: Tensor- Size([n_samples, z_dim])
Log of variance for each sample using label prior p(z | u).
-
posterior_mean
: Tensor- Size([n_samples, z_dim])
Mean for each sample using full posterior of q(z | x,u) ~ q(z | x) × p(z | u).
-
posterior_log_variance
: Tensor- Size([n_samples, z_dim])
Log of variance for each sample using full posterior of q(z | x,u) ~ q(z | x) × p(z | u).
-
z_mean
: Tensor- Size([n_samples, z_dim])
Mean for each sample using approximation of q(z | x).
-
z_log_variance
: Tensor- Size([n_samples, z_dim])
Log of variance for each sample using approximation of q(z | x).
-
z_sample
: Tensor- Size([n_samples, z_dim])
Generated latents
z
.
pi-VAE learns the deep generative model and the approximate posterior q(z | x, u) of the true posterior p(z | x, u) by maximizing the evidence lower bound (ELBO) of p(x | u).
from pi_vae_pytorch.utils import compute_loss
outputs = model(x, u) # Initialized with decoder_observation_model="poisson"
loss = compute_loss(
x=x,
firing_rate=outputs["firing_rate"],
lambda_mean=outputs["lambda_mean"],
lambda_log_variance=outputs["lambda_log_variance"],
posterior_mean=outputs["posterior_mean"],
posterior_log_variance=outputs["posterior_log_variance"],
observation_model=model.decoder_observation_model
)
loss.backward()
from pi_vae_pytorch.utils import compute_loss
outputs = model(x, u) # Initialized with decoder_observation_model="gaussian"
loss = compute_loss(
x=x,
firing_rate=outputs["firing_rate"],
lambda_mean=outputs["lambda_mean"],
lambda_log_variance=outputs["lambda_log_variance"],
posterior_mean=outputs["posterior_mean"],
posterior_log_variance=outputs["posterior_log_variance"],
observation_model=model.decoder_observation_model,
observation_noise_model=model.observation_noise_model
)
loss.backward()
-
x
: Tensor- Size([n_samples, x_dim])
Observations
x
. -
firing_rate
: Tensor- Size([n_samples, x_dim])
Predicted firing rate of generated latent
z
. -
lambda_mean
: Tensor- Size([n_samples, z_dim])
Means from label prior p(z | u).
-
lambda_log_variance
: Tensor- Size([n_samples, z_dim])
Log of variances from label prior p(z | u).
-
posterior_mean
: Tensor- Size([n_samples. z_dim])
Means from full posterior of q(z | x,u) ~ q(z | x) × p(z | u).
-
posterior_log_variance
: Tensor- Size([n_samples. z_dim])
Log of variances from full posterior of q(z | x,u) ~ q(z | x) × p(z | u).
-
observation_model
: str- One of
poisson
orgaussian
- Should use the same value passed to
decoder_observation_model
when initializingPiVAE
.
The observation model used by pi-VAE's decoder.
- One of
-
observation_noise_model
: nn.Module- Default: None
- Only applied when
observation model="gaussian"
The noise model used when pi-VAE's decoder utilizes a Gaussian observation model. When
PiVAE
is initialized withdecoder_observation_model="gaussian"
, the model'sobservation_noise_model
attribute should be used.
@misc{zhou2020learning,
title={Learning identifiable and interpretable latent models of high-dimensional neural activity using pi-VAE},
author={Ding Zhou and Xue-Xin Wei},
year={2020},
eprint={2011.04798},
archivePrefix={arXiv},
primaryClass={stat.ML}
}