Sub_Matrix_Decomposition.cpp

Go to the documentation of this file.

/*
Created on Fri Jun 26 14:13:26 2020
Copyright (C) 2020 Peter Rakyta, Ph.D.

This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program.  If not, see http://www.gnu.org/licenses/.

@author: Peter Rakyta, Ph.D.
*/
#include "qgd/Sub_Matrix_Decomposition.h"





Sub_Matrix_Decomposition::Sub_Matrix_Decomposition( Matrix Umtx_in, int qbit_num_in, bool optimize_layer_num_in=false, guess_type initial_guess_in= CLOSE_TO_ZERO ) : Decomposition_Base(Umtx_in, qbit_num_in, initial_guess_in) {

    // logical value. Set true if finding the minimum number of operation layers is required (default), or false when the maximal number of CNOT gates is used (ideal for general unitaries).
    optimize_layer_num  = optimize_layer_num_in;

    // A string describing the type of the class
    type = SUB_MATRIX_DECOMPOSITION_CLASS;

    // The global minimum of the optimization problem
    global_target_minimum = 0;

    // number of iteratrion loops in the optimization
    iteration_loops[2] = 3;

    // logical value indicating whether the quasi-unitarization of the submatrices was done or not
    subdisentaglement_done = false;

    // filling in numbers that were not given in the input
    for ( std::map<int,int>::iterator it = max_layer_num_def.begin(); it!=max_layer_num_def.end(); it++) {
        if ( max_layer_num.count( it->first ) == 0 ) {
            max_layer_num.insert( std::pair<int, int>(it->first,  it->second) );
        }
    }






}


Sub_Matrix_Decomposition::~Sub_Matrix_Decomposition() {

}






void  Sub_Matrix_Decomposition::disentangle_submatrices() {

    if (subdisentaglement_done) {
        if (verbose) {
            printf("Sub-disentaglement already done.\n");
        }
        return;
    }

    if (verbose) {
        printf("\nDisentagling submatrices.\n");
    }

    // setting the global target minimum
    global_target_minimum = 0;
    current_minimum = optimization_problem(NULL);

    // check if it needed to do the subunitarization
    if (check_optimization_solution()) {
        if (verbose) {
            printf("Disentanglig not needed\n");
        }
        subdecomposed_mtx = Umtx;
        subdisentaglement_done = true;
        return;
    }



    if ( !check_optimization_solution() ) {
        // Adding the operations of the successive layers

        //measure the time for the decompositin
        clock_t start_time = time(NULL);

        // the maximal number of layers in the subdeconposition
        int max_layer_num_loc;
        try {
            max_layer_num_loc = max_layer_num[qbit_num];
        }
        catch (...) {
            throw "Layer number not given";
        }


        // the  number of succeeding identicallayers in the subdeconposition
        int identical_blocks_loc;
        try {
            identical_blocks_loc = identical_blocks[qbit_num];
            if (identical_blocks_loc==0) {
                identical_blocks_loc = 1;
            }
        }
        catch (...) {
            identical_blocks_loc=1;
        }

        while ( layer_num < max_layer_num_loc ) {

            int control_qbit_loc = qbit_num-1;
            int solution_guess_num = parameter_num;

            for (int target_qbit_loc = 0; target_qbit_loc<control_qbit_loc; target_qbit_loc++ ) {

                for (int idx=0;  idx<identical_blocks_loc; idx++) {

                    // creating block of operations
                    Operation_block* block = new Operation_block( qbit_num );

                    // add CNOT gate to the block
                    block->add_cnot_to_end(control_qbit_loc, target_qbit_loc);

                    // adding U3 operation to the block
                    bool Theta = true;
                    bool Phi = false;
                    bool Lambda = true;
                    block->add_u3_to_end(target_qbit_loc, Theta, Phi, Lambda);
                    block->add_u3_to_end(control_qbit_loc, Theta, Phi, Lambda);

                    // adding the opeartion block to the operations
                    add_operation_to_end( block );

                }
            }

            // get the number of blocks
            layer_num = operations.size();

            // Do the optimization
            if (optimize_layer_num || layer_num >= max_layer_num_loc ) {

                // solve the optzimalization problem to find the correct mninimum
                if ( optimized_parameters == NULL ) {
                    solve_optimization_problem( optimized_parameters, 0);
                }
                else {
                    solve_optimization_problem( optimized_parameters, solution_guess_num);
                }

                if (check_optimization_solution()) {
                    break;
                }
            }

        }

        if (verbose) {
            printf("--- %f seconds elapsed during the decomposition ---\n\n", float(time(NULL) - start_time));
        }


    }



    if (check_optimization_solution()) {
        if (verbose) {
            printf("Sub-disentaglement was succesfull.\n\n");
        }
    }
    else {
        if (verbose) {
            printf("Sub-disentaglement did not reach the tolerance limit.\n\n");
        }
    }


    // indicate that the unitarization of the sumbatrices was done
    subdisentaglement_done = true;

    // The subunitarized matrix
    subdecomposed_mtx = get_transformed_matrix( optimized_parameters, operations.begin(), operations.size(), Umtx );
}





void Sub_Matrix_Decomposition::solve_layer_optimization_problem( int num_of_parameters, gsl_vector *solution_guess_gsl) {


        if (operations.size() == 0 ) {
            return;
        }


        if (solution_guess_gsl == NULL) {
            solution_guess_gsl = gsl_vector_alloc(num_of_parameters);
        }

        if (optimized_parameters == NULL) {
            optimized_parameters = (double*)qgd_calloc(num_of_parameters,sizeof(double), 64);
            memcpy(optimized_parameters, solution_guess_gsl->data, num_of_parameters*sizeof(double) );
        }

        // maximal number of iteration loops
        int iteration_loops_max;
        try {
            iteration_loops_max = std::max(iteration_loops[qbit_num], 1);
        }
        catch (...) {
            iteration_loops_max = 1;
        }


        // do the optimization loops
        for (int idx=0; idx<iteration_loops_max; idx++) {

            size_t iter = 0;
            int status;

            const gsl_multimin_fdfminimizer_type *T;
            gsl_multimin_fdfminimizer *s;

            Sub_Matrix_Decomposition* par = this;


            gsl_multimin_function_fdf my_func;


            my_func.n = num_of_parameters;
            my_func.f = optimization_problem;
            my_func.df = optimization_problem_grad;
            my_func.fdf = optimization_problem_combined;
            my_func.params = par;


            T = gsl_multimin_fdfminimizer_vector_bfgs2;
            s = gsl_multimin_fdfminimizer_alloc (T, num_of_parameters);


            gsl_multimin_fdfminimizer_set (s, &my_func, solution_guess_gsl, 0.1, 0.1);

            do {
                iter++;

                status = gsl_multimin_fdfminimizer_iterate (s);

                if (status) {
                  break;
                }

                status = gsl_multimin_test_gradient (s->gradient, 1e-1);
                /*if (status == GSL_SUCCESS) {
                    printf ("Minimum found\n");
                }*/

            } while (status == GSL_CONTINUE && iter < 100);

            if (current_minimum > s->f) {
                current_minimum = s->f;
                memcpy( optimized_parameters, s->x->data, num_of_parameters*sizeof(double) );
                gsl_multimin_fdfminimizer_free (s);

                for ( int jdx=0; jdx<num_of_parameters; jdx++) {
                    solution_guess_gsl->data[jdx] = solution_guess_gsl->data[jdx] + (2*double(rand())/double(RAND_MAX)-1)*2*M_PI/100;
                }
            }
            else {
                for ( int jdx=0; jdx<num_of_parameters; jdx++) {
                    solution_guess_gsl->data[jdx] = solution_guess_gsl->data[jdx] + (2*double(rand())/double(RAND_MAX)-1)*2*M_PI;
                }

                gsl_multimin_fdfminimizer_free (s);
            }



        }



}




double Sub_Matrix_Decomposition::optimization_problem( const double* parameters ) {

        // get the transformed matrix with the operations in the list
        Matrix matrix_new = get_transformed_matrix( parameters, operations.begin(), operations.size(), Umtx );

        double cost_function = get_submatrix_cost_function(matrix_new); //NEW METHOD


        return cost_function;
}


double Sub_Matrix_Decomposition::optimization_problem( const gsl_vector* parameters, void* void_instance ) {

    Sub_Matrix_Decomposition* instance = reinterpret_cast<Sub_Matrix_Decomposition*>(void_instance);
    std::vector<Operation*> operations_loc = instance->get_operations();

    Matrix Umtx_loc = instance->get_Umtx();
    Matrix matrix_new = instance->get_transformed_matrix( parameters->data, operations_loc.begin(), operations_loc.size(), Umtx_loc );


    double cost_function = get_submatrix_cost_function(matrix_new);  //NEW METHOD


    return cost_function;
}






int Sub_Matrix_Decomposition::set_identical_blocks( int qbit, int identical_blocks_in )  {

    std::map<int,int>::iterator key_it = identical_blocks.find( qbit );

    if ( key_it != identical_blocks.end() ) {
        identical_blocks.erase( key_it );
    }

    identical_blocks.insert( std::pair<int, int>(qbit,  identical_blocks_in) );

    return 0;

}


int Sub_Matrix_Decomposition::set_identical_blocks( std::map<int, int> identical_blocks_in )  {

    for ( std::map<int,int>::iterator it=identical_blocks_in.begin(); it!= identical_blocks_in.end(); it++ ) {
        set_identical_blocks( it->first, it->second );
    }

    return 0;

}


Sub_Matrix_Decomposition* Sub_Matrix_Decomposition::clone() {

    Sub_Matrix_Decomposition* ret = new Sub_Matrix_Decomposition(Umtx, qbit_num, optimize_layer_num, initial_guess);

    // setting computational parameters
    ret->set_identical_blocks( identical_blocks );
    ret->set_max_iteration( max_iterations );
    ret->set_optimization_blocks( optimization_block );
    ret->set_max_layer_num( max_layer_num );
    ret->set_iteration_loops( iteration_loops );

    if ( extract_operations(static_cast<Operation_block*>(ret)) != 0 ) {
        printf("Sub_Matrix_Decomposition::clone(): extracting operations was not succesfull\n");
        exit(-1);
    }

    return ret;

}