I read your MB-BAB paper with great interest. And I noticed that you formulate multi-neuron constraints as follows:
I wonder why the constant term is encoded as -p(i) instead of just p(i)?
Does it mean that the elements in p(i) are always non-negative?
I would really appreciate your clarification. Thank you!
Hi,
I read your MB-BAB paper with great interest. And I noticed that you formulate multi-neuron constraints as follows: I wonder why the constant term is encoded as -p(i) instead of just p(i)? Does it mean that the elements in p(i) are always non-negative? I would really appreciate your clarification. Thank you!