-
Word2Vec from ScratchBuilding skip-gram and CBOW word2vec models with negative sampling, training on the Stanford Sentiment Treebank.
6 min -
RBF Networks for Function ApproximationImplementing a Radial Basis Function network with K-Means clustering and LMS training to approximate a noisy sinusoidal function.
6 min -
Backpropagation and the N-Bit Parity ProblemImplementing a multi-layer perceptron from scratch to solve the N-bit parity problem, and analyzing the effects of learning rate and momentum.
5 min -
Perceptron Learning, the XOR Problem, and Gradient DescentApplying the perceptron learning rule, confronting the XOR limitation, designing multi-layer networks, and connecting gradient descent to LMS.
5 min -
McCulloch-Pitts Neurons and Linear SeparabilityDesigning M-P neurons for conditional logic, building binary adders from neural networks, and exploring the limits of linear separability.
4 min
Back