TensorFlow - l'introduction de deux L2 de régularisation et d'abandon dans le réseau. Est-il un sens?

Ok, après quelques autres efforts, j'ai réussi à le résoudre et introduire L2 et d'abandon dans mon réseau, le code est ci-dessous. J'ai eu une légère amélioration sur le même réseau sans le décrochage scolaire (avec L2 en place). Je ne sais pas si cela vaut vraiment la peine l'effort d'introduire les deux d'entre eux, L2 et d'abandon, mais au moins il travaille et s'améliore légèrement les résultats.

#ANN with introduced dropout
#This time we still use the L2 but restrict training dataset
#to be extremely small
#get just first 500 of examples, so that our ANN can memorize whole dataset
train_dataset_2 = train_dataset[:500, :]
train_labels_2 = train_labels[:500]
#batch size for SGD and beta parameter for L2 loss
batch_size = 128
beta = 0.001
#that's how many hidden neurons we want
num_hidden_neurons = 1024
#building tensorflow graph
graph = tf.Graph()
with graph.as_default():
# Input data. For the training data, we use a placeholder that will be fed
# at run time with a training minibatch.
tf_train_dataset = tf.placeholder(tf.float32,
shape=(batch_size, image_size * image_size))
tf_train_labels = tf.placeholder(tf.float32, shape=(batch_size, num_labels))
tf_valid_dataset = tf.constant(valid_dataset)
tf_test_dataset = tf.constant(test_dataset)
#now let's build our new hidden layer
#its weights
hidden_weights = tf.Variable(
tf.truncated_normal([image_size * image_size, num_hidden_neurons]))
hidden_biases = tf.Variable(tf.zeros([num_hidden_neurons]))
#now the layer itself. It multiplies data by weights, adds biases
#and takes ReLU over result
hidden_layer = tf.nn.relu(tf.matmul(tf_train_dataset, hidden_weights) + hidden_biases)
#add dropout on hidden layer
#we pick up the probabylity of switching off the activation
#and perform the switch off of the activations
keep_prob = tf.placeholder("float")
hidden_layer_drop = tf.nn.dropout(hidden_layer, keep_prob)  
#time to go for output linear layer
#out weights connect hidden neurons to output labels
#biases are added to output labels  
out_weights = tf.Variable(
tf.truncated_normal([num_hidden_neurons, num_labels]))  
out_biases = tf.Variable(tf.zeros([num_labels]))  
#compute output
#notice that upon training we use the switched off activations
#i.e. the variaction of hidden_layer with the dropout active
out_layer = tf.matmul(hidden_layer_drop,out_weights) + out_biases
#our real output is a softmax of prior result
#and we also compute its cross-entropy to get our loss
#Notice - we introduce our L2 here
loss = (tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(
out_layer, tf_train_labels) +
beta*tf.nn.l2_loss(hidden_weights) +
beta*tf.nn.l2_loss(hidden_biases) +
beta*tf.nn.l2_loss(out_weights) +
beta*tf.nn.l2_loss(out_biases)))
#now we just minimize this loss to actually train the network
optimizer = tf.train.GradientDescentOptimizer(0.5).minimize(loss)
#nice, now let's calculate the predictions on each dataset for evaluating the
#performance so far
# Predictions for the training, validation, and test data.
train_prediction = tf.nn.softmax(out_layer)
valid_relu = tf.nn.relu(  tf.matmul(tf_valid_dataset, hidden_weights) + hidden_biases)
valid_prediction = tf.nn.softmax( tf.matmul(valid_relu, out_weights) + out_biases) 
test_relu = tf.nn.relu( tf.matmul( tf_test_dataset, hidden_weights) + hidden_biases)
test_prediction = tf.nn.softmax(tf.matmul(test_relu, out_weights) + out_biases)
#now is the actual training on the ANN we built
#we will run it for some number of steps and evaluate the progress after 
#every 500 steps
#number of steps we will train our ANN
num_steps = 3001
#actual training
with tf.Session(graph=graph) as session:
tf.initialize_all_variables().run()
print("Initialized")
for step in range(num_steps):
# Pick an offset within the training data, which has been randomized.
# Note: we could use better randomization across epochs.
offset = (step * batch_size) % (train_labels_2.shape[0] - batch_size)
# Generate a minibatch.
batch_data = train_dataset_2[offset:(offset + batch_size), :]
batch_labels = train_labels_2[offset:(offset + batch_size), :]
# Prepare a dictionary telling the session where to feed the minibatch.
# The key of the dictionary is the placeholder node of the graph to be fed,
# and the value is the numpy array to feed to it.
feed_dict = {tf_train_dataset : batch_data, tf_train_labels : batch_labels, keep_prob : 0.5}
_, l, predictions = session.run(
[optimizer, loss, train_prediction], feed_dict=feed_dict)
if (step % 500 == 0):
print("Minibatch loss at step %d: %f" % (step, l))
print("Minibatch accuracy: %.1f%%" % accuracy(predictions, batch_labels))
print("Validation accuracy: %.1f%%" % accuracy(
valid_prediction.eval(), valid_labels))
print("Test accuracy: %.1f%%" % accuracy(test_prediction.eval(), test_labels))

En fait, le document original utilise max-norme de la régularisation et de ne pas en L2, en plus d'abandon:
"Le réseau de neurones a été optimisé sous la contrainte ||w||2 ≤ c. Cette
la contrainte a été imposée lors de l'optimisation, par la projection de w sur la surface d'une boule de rayon c, lorsque w est allé hors de lui. Il est également appelé max-norme de régularisation, puisqu'elle implique que la valeur maximale que la norme de tout le poids est c" (http://jmlr.org/papers/volume15/srivastava14a/srivastava14a.pdf)

Vous pouvez trouver une belle discussion à propos de cette méthode de régularisation ici: https://plus.google.com/+IanGoodfellow/posts/QUaCJfvDpni