Neural Network Model (NNet) Training Task

A Neural Network Model (NNet) Training Task is a model-based training task that is a neural network creation task (that can produce an artificial neural network).

• Context:
  • output: Neural Network Model Instance.
  • It can be solved by a ANN Training System (that implements an NNet training algorithm).
  • It can range from being Single-Layer Neural Network Training to being Multi-Layer Neural Network Training.
  • It can range from being an ANN Classifier Training Task to being an ANN Ranker Training Task to being an ANN Estimator Training Task.
  • It can range from (typically) being a Data-Driven Neural Network Training Task to being a Heuristic Neural Network Training Task.
• Example(s):
  • a Feed-Forward NNet Training Task, such as a [CCN training task]] for an RNN training task.
  • a Language Model NNet Training, Image Classifier NNet Training, ....
  • …
• Counter-Example(s):
  • SVM Training, Logistic Training, Decision Tree Training, Hidden Markov Model Training, Conditional Random Fields Model Training, ...
  • Neural Network Model Inference Task.
• See: CNN Model Training.

---

---

References

2014

• http://deeplearning.stanford.edu/wiki/index.php/Backpropagation_Algorithm
  • QUOTE: Suppose we have a fixed training set [math]\displaystyle{ \{ (x^{(1)}, y^{(1)}), \ldots, (x^{(m)}, y^{(m)}) \} }[/math] of [math]\displaystyle{ m }[/math] training examples. We can train our neural network using batch gradient descent ...

---