No subject

feed-forward neural network. From this,
I suggest that you take a step back and go to a vector/matrix
representation.

This will reduces a lot of the complexity in your codes.

Moreover, many of the advanced training algorithms (Conjugate Gradient,
BFGS, L-BFGS) requires the representation
of the weight parameters to be in vector form. If the HNN project continues
to use a graph representation,
it would be costly to convert back-and-forth from graph to vector to use
the advance training algorithms.

I understand that you guys want to make the most general representation
possible so the project would be scalable, but I think
it would behoove you guys to re-evaluate your representation of the neural
network.

Thanks.

Kiet Lam

--0016e6de002781539804b66d227f
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Hey, I don't know how active the HNN project is, but after looking at y=
our code draft, I have a few suggestions.<div><br></div><div>You guys are m=
aking it way too hard on yourselves by going to a graph representation.=A0<=
/div>
<div>This will no doubt have a performance impact and will / already has in=
troduced needless complexity to your codes.</div><div><br></div><div>From t=
he Wiki page of HNN, it seems like you are interested in a feed-forward neu=
ral network. From this,</div>
<div>I suggest that you take a step back and go to a vector/matrix represen=
tation.</div><div><br></div><div>This will reduces a lot of the complexity =
in your codes.</div><div><br></div><div>Moreover, many of the advanced trai=
ning algorithms (Conjugate Gradient, BFGS, L-BFGS) requires the representat=
ion</div>
<div>of the weight parameters to be in vector form. If the HNN project cont=
inues to use a graph representation,</div><div>it would be costly to conver=
t back-and-forth from graph to vector to use the advance training algorithm=
s.</div>
<div><br></div><div>I understand that you guys want to make the most genera=
l representation possible so the project would be scalable, but I think</di=
v><div>it would behoove you guys to re-evaluate your representation of the =
neural network.</div>
<div><br></div><div>Thanks.</div><div><br></div><div>Kiet Lam</div>

--0016e6de002781539804b66d227f--