1. PANN Testing

2. Training Time and Error Benchmarking Target error: 0.02 (Mean Squared Error)

3. Training Time and Error Benchmarking • Progress ANN reduces its training error to desired minimum in less than one second. • Perceptron and Back Propagation ANNs have substantial error, which reduces slowly. Perceptron and Back Propagation ANNs showno tendency to reach target error. • Tested training set: • 30,000 images 60 Perceptron 50 40 Mean Squared Error (x 100) 30 Back propagation 20 Progress ANN 10 Target error: 0.02 0 50 250 1250 6250 31250 156250 781250 Time (milliseconds)

4. Additional Comparison to Alternatives Progress P-network NeuroSolutions Progress P-Net Creator NeuroSolution Data Manager

5. Additional Comparison to Alternatives Number of: records, images, lines, samples, data set and sample data are used synonymously. Comparison of training time between IBM SPSS Statistics 22andP-network, for the same problem, tested on Apple iMac 27" 3.5 GHz quad-core Intel Core i7 8GB of 1600MHz DDR3 memory; SSD 14000 • IBM SPSS Statistics 22 • Exponential growth of training time 12000 10000 8000 Time in seconds 6000 4000 2000 Training advantage factor = Numberof images 0 1000 2000 3000 4000 5000 6000 7000 0 • ProgressP-network • Linear growth of training time 4 3 2 Time in seconds 1 Numberof images 0 1000 2000 3000 4000 5000 6000 7000 0

6. PANN vs. Classical Neural Network • Theoretically unlimited intelligence • Practically unlimited intelligence Classical neural network: PANN Human brain • Long training time • Quick training Network Intelligence is proportional to the number of elements and problems at hand 1 hour 1 month 1 thousand years 1 million years 1 minute 24 hour 1 year PANN requires minutes (at most - hours) to reach the level of network intelligence unreachable for classical neural networks for thousands of years Network Intelligence Classical neural network PANN Hardware PANN Software Seconds 0 10 102 103 104 105 106 107 108 109 1010 1011 1012 1013 1014 Training Time

7. Test on images compression Files with images from CIFAR-10 test site No compression 20000 MSE:0.006 MSE:0.013 MSE:0.005 15000 MSE:0.011 Economy of disk space Number of color images in a file 10000 MSE:0.008 MSE:0.004 PANN with 50 weights on a synapse PANN with 100 weights on a synapse 5000 1000 100 1 50 100 150 200 250 300 File size on disk, MB

8. #2 GPUs breakthrough - 2 Amdahl’s Law and P-net PANN simple matrix algebra mathematics allows 100% parallel processing. Thus, speed increases linearly with additional GPUs and CPUs. Progress US patent application 15/449,614 that covers matrix algebra application with PANN, is filed on March 3, 2017. • This allows building computers and other electronics with: • very high processing speed, and • reduced number of GPUs and CPUs P-net application

9. #2 GPUs breakthrough - 2 Trading speed: Comparison of PANN and nVidia cuDNN PANN training speed on CPU and GPUs is thousand times higher than of existing ANN. • Allows to: • Improve ANN training speed thousands times • Build supercomputer on GPUs • Build Hypercomputer on GPUs PANNprovides 60 (threads on GPU) x = 201 000 x. Acceleration is proportional to the number (N) of GPUs: 201 000 xN

10. Comparison of P-net with CPU/GPU CPU - central processing unit GPU - graphics processing unit 1 sec = 1000 msec Testing computer with CPU speed/ GPU speed = 4 time, log

11. Comparison of P-net with CPU/GPU CPU - central processing unit GPU - graphics processing unit time, log