Question:
What do you mean by Attention? What are the types of Attention in Neural Networks?
Expected Answer:
Define attention as a representation of a distribution learnt by a Neural Network, use case and an example. You could also specify types of attention and give applications of each.
Neural attention mechanism equips a neural network with the ability to focus on a subset of its inputs (or features): it selects specific inputs. Let be an input vector, a feature vector, an attention vector, an attention glimpse and an attention network with parameters . Typically, attention is implemented as
where is element-wise multiplication, while is an output of another neural network with parameters . We can talk about soft attention, which multiplies features with a (soft) mask of values between zero and one, or hard attention, when those values are constrained to be exactly zero or one, namely . In the latter case, we can use the hard attention mask to directly index the feature vector: , which changes its dimensionality and now with .
Types of Attention:
1. Visual
2. Hard Attention
3. Soft Attention
4. Gaussian Attention
Read more:
Comments
Post a Comment