On: Learning to Learn in TensorFlow

[_pctq]

From the related article:

> The move from hand-designed features to learned features in machine learning has been wildly successful.

Are the features here the "feature vectors" or the network architecture? Or something else? In other terms, does this project help normalizing data, or does it help tweaking hyper parameters?

[conmdur → _pctq]

Here the features are the feature vectors themselves, yes. It's been found that taking somewhat of a hands-off approach and allowing networks to engineer their own mid-level representations from raw data can be very beneficial.

This is the idea behind the learning to learn paper. Instead of taking our gradient and plugging it in to a hand-engineered (i.e. on paper) update rule, we feed it to a neural network, which is trained to find the optimal update rule, in some sense (neural networks are just function approximators after all).

[dougabug → _pctq]

The point of the original paper was to learn the hyperparameters of a DNN using a DNN, as opposed to using, say, a bayesian optimization framework.