# RelaxedOneHotCategorical in Edward

#1

I am using RelaxedOneHotCategorical in Edward model, similar to

logits = nn(..)
tau = Gamma(0.1, 0.5)
y_prob = RelaxedOneHotCategorical(tau, logits)


and use KLqp for inference, but my Loss is negative after few iterations. My question is

a) Is RelaxedOneHotCategorical supported in Edward
b) Is my usage correct?
c) Are there any issues with KLqp inference when using such a model?

#2

a) Is RelaxedOneHotCategorical supported in Edward?

b) Is my usage correct?

I think that modeling of temperature parameters may not be appropriate.
What about trying to set the temperature parameter as a constant?
For example,
tau = tf.constant(0.5)
In the first experiment, they used a fixed \tau=1.
\tau = max(0.5, exp(−rt)) of the global training step t.