Feedback Layer relevancy of Neural Network
1. What is the Feedback Layer ?
Imagine we’re learning how to shoot a basketball.
- We throw the ball.
- If it goes in, we feel good.
- If it misses, we adjust how we throw the next time.
That “adjustment” is based on feedback — and that’s exactly what happens in a Feedback Layer of a neural network.
In the context of a neural network:
- The feedback layer is not a physical layer like input or hidden layers.
- It’s part of the learning loop.
- After the network makes a prediction, it compares that prediction to the correct answer.
- Then, it calculates how wrong it was (this is called loss or error).
- This error is sent backward through the network to adjust the weights — that process is called backpropagation.
- So, the feedback layer helps the network learn from its mistakes.
Analogy:
- Neural network = Student
- Output = Answer on the test
- Feedback layer = Teacher correcting the answer
- Backpropagation = Student learning from the mistake and improving next time
So, what does the feedback layer do?
Step | What Happens |
---|---|
1 | Neural net gives an answer (output) |
2 | We check how far off that answer is (calculate error) |
3 | We send that error back (feedback) to update the thinking process (weights) |
4 | Neural net improves for next time |
Feedback Layer relevancy of Neural Network – Feedback Layer example with Simple Python