Neural Networks

An introduction to neural networks covering a brief look at a how they work and in what scenarios they are useful.

Transcript

1. Neural Networks Gunther Cox
2. Why Neural Networks? • Noisy data • Fault tolerance • Adapting to circumstances • Massively parallel
4. 1 if ... W 0 * I 0 + W 1 * I 1 + W b > 0 0 if ... W 0 * I 0 + W 1 * I 1 + W b <= 0
5. I 0 I 1 Desired output 0 0 0 0 1 1 1 0 1 1 1 1
6. double activation = 0; for (int i=0; i < n; i++) { activation += x[i] * w[i] * b[i]; }
7. Feedforward design
8. Applications of Neural Networks • Aircraft engines • Stock market • Pharmacy • Banking • Google and NVIDIA
9. Licensed CC Attribution 3.0 Gunther Cox 2014 http://salviusrobot.blogspot.com