This programming exercise was done as part of Coursera's Machine Learning Course (Stanford University), taught by Prof. Andrew Ng.
- Built a model to recognize handwritten digits using neural networks.
- Implemented backpropagation to compute gradients for the parameters of the neural network, trained the neural network using Octave's fmincg optimization solver and used feedforward propagation to make predictions.
- Computed training accuracy for the neural network, studied the effect of varying learning settings on the training accuracy