Presenting: Itai Avron Supervisor: Chen Koren Characterization Presentation Spring 2005 Implementation of Artificial Intelligence System on FPGA
Project Goals Creating a VHDL design of a Neural Network Comparison Vs. software implementation (Matlab) If the time will suffice: Implementing Learning Algorithm (BP)
Background Neural Network is a Learning Machine It is build from Neurons (Perceptrons), which holds the knowledge of the system within their inter-connection strength Every Neuron Implement the Active Function:
System Interface Input: - Image (16x16 pixels) - Weights Output: - A number between 0-9 (4 bit vector)
System Architecture control SRAM (Xilinx) Neuron Input Output …
Neuron Architecture Multipliers for calculating (Xilinx) SRAM implementing Activation Function SRAM saving W and X
Controller – Flow Diagram Idle Upload Image and Weights Calculate Hidden Layer Calculate Exit Layer New Image Until all Neurons are calculated Upload Input to Neuron Neuron Calculation Save Neuron Result
Schedule : Planning the Controller, Block Diagram : Implementing the NN in Matlab : Implementing a single Neuron : Implementing the controller End of July: Midterm Presentation : Implementing the NN : Debugging and Simulating the NN September: Final Presentation