Extra Nengo objects

NengoDL adds some new Nengo objects that can be used during model construction. These could be used with any Simulator, not just nengo_dl, but they tend to be useful for deep learning applications.

Neuron types

Additions to the neuron types included with Nengo.

class nengo_dl.neurons.SoftLIFRate(sigma=1.0, **lif_args)

LIF neuron with smoothing around the firing threshold.

This is a rate version of the LIF neuron whose tuning curve has a continuous first derivative, due to the smoothing around the firing threshold. It can be used as a substitute for LIF neurons in deep networks during training, and then replaced with LIF neurons when running the network [1].

Parameters:

sigma : float

Amount of smoothing around the firing threshold. Larger values mean more smoothing.

tau_rc : float

Membrane RC time constant, in seconds. Affects how quickly the membrane voltage decays to zero in the absence of input (larger = slower decay).

tau_ref : float

Absolute refractory period, in seconds. This is how long the membrane voltage is held at zero after a spike.

Notes

Adapted from https://github.com/nengo/nengo_extras/blob/master/nengo_extras/neurons.py

References

[1]	(1, 2) Eric Hunsberger and Chris Eliasmith (2015): Spiking deep networks with LIF neurons. https://arxiv.org/abs/1510.08829.

rates(x, gain, bias)

Always use LIFRate to determine rates.

step_math(dt, J, output)

Compute rates in Hz for input current (incl. bias)

Distributions

Additions to the distributions included with Nengo. These distributions are usually used to initialize weight matrices, e.g. nengo.Connection(a.neurons, b.neurons, transform=nengo_dl.dists.Glorot()).

class nengo_dl.dists.TruncatedNormal(mean=0, stddev=1, limit=None)

Normal distribution where any values more than some distance from the mean are resampled.

Parameters:

mean : float, optional

mean of the normal distribution

stddev : float, optional

standard deviation of the normal distribution

limit : float, optional

resample any values more than this distance from the mean. if None, then limit will be set to 2 standard deviations

sample(n, d=None, rng=None)

Samples the distribution.

Parameters:

n : int

Number samples to take.

d : int or None, optional

The number of dimensions to return. If this is an int, the return value will be of shape (n, d). If None, the return value will be of shape (n,).

rng : numpy.random.RandomState, optional

Random number generator state (if None, will use the default numpy random number generator).

Returns:

samples : (n,) or (n, d) array_like

Samples as a 1d or 2d array depending on d. The second dimension enumerates the dimensions of the process.

class nengo_dl.dists.VarianceScaling(scale=1, mode='fan_avg', distribution='uniform')

Variance scaling distribution for weight initialization (analogous to TensorFlow init_ops.VarianceScaling).

Parameters:

scale : float, optional

overall scale on values

mode : “fan_in” or “fan_out” or “fan_avg”, optional

whether to scale based on input or output dimensionality, or average of the two

distribution: “uniform” or “normal”, optional

whether to use a uniform or normal distribution for weights

sample(n, d=None, rng=None)

Samples the distribution.

Parameters:

n : int

Number samples to take.

d : int or None, optional

The number of dimensions to return. If this is an int, the return value will be of shape (n, d). If None, the return value will be of shape (n,).

rng : numpy.random.RandomState, optional

Random number generator state (if None, will use the default numpy random number generator).

Returns:

samples : (n,) or (n, d) array_like

Samples as a 1d or 2d array depending on d. The second dimension enumerates the dimensions of the process.

class nengo_dl.dists.Glorot(scale=1, distribution='uniform')

Weight initialization method from [1] (also known as Xavier initialization).

Parameters:

scale : float, optional

scale on weight distribution. for rectified linear units this should be sqrt(2), otherwise usually 1

distribution: “uniform” or “normal”, optional

whether to use a uniform or normal distribution for weights

References

[1]	(1, 2) Xavier Glorot and Yoshua Bengio (2010): Understanding the difficulty of training deep feedforward neural networks. International conference on artificial intelligence and statistics. http://proceedings.mlr.press/v9/glorot10a/glorot10a.pdf.

class nengo_dl.dists.He(scale=1, distribution='normal')

Weight initialization method from [1].

Parameters:

scale : float, optional

scale on weight distribution. for rectified linear units this should be sqrt(2), otherwise usually 1

distribution: “uniform” or “normal”, optional

whether to use a uniform or normal distribution for weights

References

[1]	(1, 2) Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. (2015): Delving deep into rectifiers: Surpassing human-level performance on ImageNet classification. https://arxiv.org/abs/1502.01852.