New answers tagged

1

The answer to your first question is because the line 'update the critic by minimising the loss $L = \frac{1}{N} \sum_i \left( y_i - Q(s_i, a_i |\theta^Q)\right)^2$ is implying that you will do this by using a gradient, i.e. you calculate the gradient of the loss wrt the parameters and perform a gradient descent step. For the second question, I am not 100% ...


0

You could just do this; concatenate your input_vector with zero's vector that has the size of your output. Then in the first pass you concatenate with the output instaid of the zero's vector. After that repeat.. At the end just compare (compute the loss) your entire output from t0 to t1 to your target and backprop. You might want to look into recurrent ...


1

You want to look at recurrent neural networks.


2

Regularizer's are used as a means to combat over fitting.They essentially create a cost function penalty which tries to prevent quantities from becoming to large. I have primarily used kernel regularizers. First I try to control over fitting using dropout layers. If that does not do the job or leads to poor training accuracy I try the Kernel regularizer. I ...


0

From my personal experience, the units hyperparam in LSTM is not necessary to be the same as max sequence length. Add more units to have the loss curve dive faster. And about the number of LSTM layers, trying out a single LSTM layer is a good start point, the model trains better with more LSTM layers. For example, MAX_SEQ_LEN=10, in Keras: Lstm1 = LSTM(units=...


0

If there is some correlation between features, that is what the network will ideally find out on its own and learn to utilize. So, in general, don't take correlated samples or features out of the training loop only because they look correlated. After all, they could convey a lot of valuable information. When it comes to correlation between data samples ...


0

Since you have a multiclass classification problem rather than a binary classification problem (i.e. a two-class problem), I recommend to adjust your architecture and use softmax instead of sigmoid as final activation function and categorcal_crossentropy instead of binary_crossentropy. Softmax will ensure all your outputs are valid probabilities. This is ...


Top 50 recent answers are included