Backpropagation Algorithm 2006 11 13 692021 1 Perceptron
Backpropagation Algorithm 郝红侠 2006. 11. 13 6/9/2021 1
Perceptron : Single Layer Feedforward Rosenblatt’s Perceptron: a network of processing elements (PE): Input layer of source nodes 6/9/2021 Output layer of neurons 2
Perceptron : Multi Layer Feedforward • • • • Input layer Output layer • • Hidden Layer 6/9/2021 4
Learning Rule Measure error l Reduce that error l l 6/9/2021 By appropriately adjusting each of the weights in the network 8
BP Network Details l Forward Pass: l l l Error is calculated from outputs Used to update output weights Backward Pass: l l Error at hidden nodes is calculated by back propagating the error at the outputs through the new weights Hidden weights updated 6/9/2021 9
Case 2: 隐层权值修正 nj aj δj 6/9/2021 wji ni wji i δi 12
THANK YOU! 6/9/2021 17
- Slides: 17