Timeline for A challenging question of ANN
Current License: CC BY-SA 4.0
14 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 12, 2021 at 10:39 | comment | added | Igor F. | @Aksakal Even then, I cannot follow. Can you please give an exact, unambiguous definition of your three-state neuron, and the weights and biases for the network E? | |
| Jan 11, 2021 at 20:06 | comment | added | Aksakal | @IgorF. updated my answer. For some reason I took it that the bipolar neuron has a ground state 0, i.e. three outputs -1,0 and 1. Clearly, everyone else considers bipolar a neuron with only two outputs -1 and 1. that's the reason my answer was confusing, me thinks | |
| Jan 11, 2021 at 20:04 | history | edited | Aksakal | CC BY-SA 4.0 |
added 206 characters in body
|
| Jan 11, 2021 at 19:54 | comment | added | Aksakal | @kevin307505 sign and step are the same. i understand confusion with my answer now: by bipolar I mean a neuron with three outputs: -1,0,1 | |
| Jan 7, 2021 at 20:49 | history | edited | Aksakal | CC BY-SA 4.0 |
pca removed
|
| Jan 7, 2021 at 20:47 | comment | added | Aksakal | @IgorF. got your point on PCA vs rotation. I'll clarify my answer. let me think of the weights question | |
| Jan 7, 2021 at 20:44 | comment | added | Igor F. | @Aksakal Rotation is a part of PCA (after identifying the principal components), but not every rotation is PCA. In our case, the problem is that no component is 'principal': the dataset is, for all we can see, rotation-symmetric. So your rotation is arbitrary, not based on PCs. On a different topic: Would you mind providing the weights and biases? | |
| Jan 7, 2021 at 20:10 | comment | added | Igor F. | @Aksakal Would you mind providing the weights and the biases for the network (E) which would solve the problem? Re. Sideways: PCA is of no help here, as there is no axis along which the data vary more than along any other. What you're doing is simple rotation (a linear operation, which you can do in the network using an additional layer). Your solution, sgn(S*V), requires the output neuron to do multiplication, but the neurons in question are only allow to sum their inputs. | |
| Jan 7, 2021 at 16:04 | comment | added | Aksakal | @DaviedZuhraph yes, looks like it. the way to think of this is that the problem is to split the area into 4 parts then combine them into two relevant parts for the answer. so first layer creates south-west and north-east. then second layer first neuron creates west and the rest, while the second neuron creates east or the rest. finally the third neuron picks east and west for L, or the opposite is not L | |
| Jan 7, 2021 at 15:54 | history | edited | Aksakal | CC BY-SA 4.0 |
edited body
|
| Jan 7, 2021 at 15:53 | comment | added | Aksakal | with what you call a step function, all you could do is criss-cross, i.e. two neurons. there's no advantage in adding more neurons in first layer. you need to multiple next. so you save your neurons for the next layer where you do it. once you multiplied, the third and final layer can pick the Ls. that's why D is the answer. | |
| Jan 7, 2021 at 15:49 | history | edited | Aksakal | CC BY-SA 4.0 |
added 85 characters in body
|
| Jan 7, 2021 at 15:37 | history | edited | Aksakal | CC BY-SA 4.0 |
added 336 characters in body
|
| Jan 7, 2021 at 15:31 | history | answered | Aksakal | CC BY-SA 4.0 |