WEBVTT

00:02.100 --> 00:06.240
Hello and welcome to this new section, recurrent neural network.

00:06.600 --> 00:14.220
Let us begin a recurrent neural network is a class of artificial neural network in Veach.

00:14.550 --> 00:18.930
The neurons are able to send feedback signals to each other.

00:19.950 --> 00:28.020
A recurrent neural network has a memory which remembers all the information about Vork has been calculated

00:28.740 --> 00:29.700
in simple words.

00:29.730 --> 00:36.300
We can say that the recurrent neural networks are the artificial neural networks with the ability to

00:36.300 --> 00:38.370
send feedback signals.

00:41.520 --> 00:49.810
And in our CNN excepts of fixed size of Victor as an input and produce a fixed size vector as an output.

00:51.330 --> 00:59.430
Whereas adenine allow us to operate over a vector sequences that in sequences, in input sequences,

00:59.490 --> 01:04.440
in output, all sequences in both input and output.

01:07.550 --> 01:13.430
As you can see here in this picture, this is a single neuron in recurrent neural network.

01:14.210 --> 01:17.150
No, no, down at the unfolded side of neuron.

01:17.420 --> 01:19.520
These are not three different neurons.

01:19.760 --> 01:21.360
It is a single neuron.

01:21.590 --> 01:23.720
But the timestamp is different.

01:24.230 --> 01:32.500
So these are the three times stands B minus one, B and D plus one after folding these three timestamps.

01:32.570 --> 01:35.990
We get a neuron like this at the left hand side.

01:36.710 --> 01:41.820
So this way, a single neuron is transmitting its output to its input.

01:42.230 --> 01:45.170
And then an entire network does that.

01:45.290 --> 01:48.070
Then it is called as recurrent neural network.

01:51.070 --> 01:53.320
So this is the recurrent neural network.

01:53.590 --> 01:58.110
This network is giving back its output to the network again.

01:58.960 --> 02:02.770
So this is all about the introduction to the current neural network.

02:03.040 --> 02:06.310
I will see you in the next it odille till then.

02:06.520 --> 02:07.500
Happy learning.