Contents

- 1 How do you handle situation where neural network is trapped in local minima?
- 2 How do you avoid local minima in neural networks?
- 3 How does TensorFlow work with the XOR problem?
- 4 How is a XOR gate created in a neural network?
- 5 What are the learned parameters of a neural network?
- 6 What kind of neural network does TensorFlow use?

## How do you handle situation where neural network is trapped in local minima?

Neural network seems stuck in a local minimum, no improvements

- import library/packages.
- load data.
- split x and y data.
- split training and test data.
- generate vector result for output – one hot style.
- combine the input and results back for use in the network.
- create the network class.
- build my network.

## How do you avoid local minima in neural networks?

However, weight adjusting with a gradient descent may result in the local minimum problem. Repeated training with random starting weights is among the popular methods to avoid this problem, but it requires extensive computational time.

## How does TensorFlow work with the XOR problem?

TensorFlow will automatically fill them with the data when we run the network. In our XOR problem, we have four different training examples and each example has two features. There are also four expected outputs, each with just one value (either a 0 or 1). In TensorFlow, this looks like this:

## How is a XOR gate created in a neural network?

The XOR gate can be created by the following combination of a NOT AND gate and an OR gate (from blog.abhranil.net) The “knowledge” of a neural network is all contained in the learned parameters which are the weights and bias.

## What are the learned parameters of a neural network?

The “knowledge” of a neural network is all contained in the learned parameters which are the weights and bias. The weights are multiplied to each signal sent by their respective perceptrons and the bias are added as y ( x) = w x + b where w is the weight and b is the bias.

## What kind of neural network does TensorFlow use?

Last November, Google open sourced the machine learning library that it uses within their own products. There are two API’s; one for Python and one for C++. Naturally, it makes sense to see what TensorFlow would make of the same network that we previously looked at and compare both Python-based neural networks.