/** Copyright (c) 2013, the SemanticVectors AUTHORS. All rights reserved. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: * Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. * Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. * Neither the name of the University of Pittsburgh nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. **/ package pitt.search.semanticvectors.vectors; import java.util.ArrayList; import java.util.Random; import pitt.search.semanticvectors.FlagConfig; public class SemanticVectorCollider { public static void main(String[] args) { FlagConfig flagConfig = FlagConfig.getFlagConfig(args); args = flagConfig.remainingArgs; Random random = new Random(); int iterations = 1000; //number of times to perform experiment int superpositions = 15000; //number of superpositions per experiment (at most) int min = Integer.MAX_VALUE; int max = Integer.MIN_VALUE; System.out.println("Number of iterations "+iterations); System.out.println("Number of superpositions per iteration (if no collision occurs) "+superpositions); System.out.println("Vector type "+flagConfig.vectortype()); System.out.println("Dimension "+flagConfig.dimension()); System.out.println("Seed length "+flagConfig.seedlength()); int overlapcnt = 0; int overlaprank = 0; ArrayList<Double> overlapRank = new ArrayList<Double>(); int overlapcount = 0; double overlapscore=0; ArrayList<Double> overlapScore = new ArrayList<Double>(); for (int cnt = 0; cnt < iterations; cnt++) { System.err.println("\nIteration "+cnt); Vector originalVector = VectorFactory.generateRandomVector( flagConfig.vectortype(), flagConfig.dimension(), flagConfig.seedlength(), random); Vector superPosition = VectorFactory.createZeroVector(flagConfig.vectortype(), flagConfig.dimension()); superPosition.superpose(originalVector, 1, null); if (flagConfig.vectortype() == VectorType.BINARY) { ((BinaryVector) superPosition).tallyVotes(); } Vector additionalVector = VectorFactory.generateRandomVector( flagConfig.vectortype(), flagConfig.dimension(), flagConfig.seedlength(), random); for (int x =0; x < superpositions; x++) { if (x % 100 == 0) System.err.print("..."); double overlapWithOrigin = superPosition.measureOverlap(originalVector); //generate another random vector Vector randomVector = VectorFactory.generateRandomVector( flagConfig.vectortype(), flagConfig.dimension(), flagConfig.seedlength(), random); double overlapWithRandom = superPosition.measureOverlap(randomVector); overlapscore += overlapWithRandom; overlapScore.add(new Double(overlapWithRandom)); if (overlapWithRandom >= overlapWithOrigin) //version 2.0 based on Roger Schvaneveldt's Matlab edition: compare superposition:origin vs. superposition:random (this is different than the implementation in Wahle et al 2012, which compared origin:superposition vs. origin:random) { System.out.println("Iteration " +cnt+": Incidental overlap occurred at superposition number "+x); min = Math.min(min,x); max = Math.max(max,x); overlapcnt++; overlaprank += x; overlapRank.add(new Double(x)); x = 999999999; } additionalVector = VectorFactory.generateRandomVector( flagConfig.vectortype(), flagConfig.dimension(), flagConfig.seedlength(), random); superPosition.superpose(additionalVector, 1, null); if (flagConfig.vectortype() == VectorType.BINARY) { ((BinaryVector) superPosition).tallyVotes(); } } } double stdRank = calculateSTD(overlapRank, (double) overlaprank/ (double) overlapcnt); System.out.println("Collisions occurred in "+(100)*((double) overlapcnt/(double) iterations) +"% of iterations"); System.out.println("\nAverage collision rank "+ (double) overlaprank/ (double) overlapcnt); System.out.println("STD collision rank "+ stdRank); System.out.println("Minimum collision rank "+min); System.out.println("Maximum collision rank "+max); } public static double calculateSTD(ArrayList<Double> values, double mean) { double std = 0; for (int x=0; x < values.size(); x++) { std += Math.pow(values.get(x).doubleValue() - mean,2); } std = std/(double) values.size()-1; std = Math.sqrt(std); return std; } }