SingleFITCLaplacianInferenceMethod.h

Go to the documentation of this file.

/*
 * Copyright (c) The Shogun Machine Learning Toolbox
 * Written (W) 2015 Wu Lin
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *
 * 1. Redistributions of source code must retain the above copyright notice, this
 *    list of conditions and the following disclaimer.
 * 2. 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.
 *
 * 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.
 *
 * The views and conclusions contained in the software and documentation are those
 * of the authors and should not be interpreted as representing official policies,
 * either expressed or implied, of the Shogun Development Team.
 *
 */

#ifndef CSINGLEFITCLAPLACIANINFERENCEMETHOD_H
#define CSINGLEFITCLAPLACIANINFERENCEMETHOD_H

#include <shogun/lib/config.h>

#ifdef HAVE_EIGEN3

#include <shogun/machine/gp/SingleFITCLaplacianBase.h>

namespace shogun
{

class CSingleFITCLaplacianInferenceMethod: public CSingleFITCLaplacianBase
{
friend class CFITCPsiLine;
public:
    CSingleFITCLaplacianInferenceMethod();

    CSingleFITCLaplacianInferenceMethod(CKernel* kernel, CFeatures* features,
            CMeanFunction* mean, CLabels* labels, CLikelihoodModel* model,
            CFeatures* inducing_features);

    virtual ~CSingleFITCLaplacianInferenceMethod();

    virtual const char* get_name() const { return "SingleFITCLaplacianInferenceMethod"; }


    virtual EInferenceType get_inference_type() const { return INF_FITC_LAPLACIAN_SINGLE; }

    static CSingleFITCLaplacianInferenceMethod* obtain_from_generic(CInferenceMethod* inference);

    virtual bool supports_regression() const
    {
        check_members();
        return m_model->supports_regression();
    }

    virtual bool supports_binary() const
    {
        check_members();
        return m_model->supports_binary();
    }

    virtual SGVector<float64_t> get_diagonal_vector();

    virtual SGVector<float64_t> get_posterior_mean();

    virtual SGMatrix<float64_t> get_posterior_covariance();

    virtual float64_t get_newton_tolerance() { return m_tolerance; }

    virtual void set_newton_tolerance(float64_t tol) { m_tolerance=tol; }

    virtual int32_t get_newton_iterations() { return m_iter; }

    virtual void set_newton_iterations(int32_t iter) { m_iter=iter; }

    virtual float64_t get_minimization_tolerance() { return m_opt_tolerance; }

    virtual void set_minimization_tolerance(float64_t tol) { m_opt_tolerance=tol; }

    virtual float64_t get_minimization_max() { return m_opt_max; }

    virtual void set_minimization_max(float64_t max) { m_opt_max=max; }

    virtual void update();

    virtual float64_t get_negative_log_marginal_likelihood();
protected:
    virtual void compute_gradient();

    virtual void update_init();

    virtual void update_alpha();

    virtual void update_chol();

    virtual void update_approx_cov();

    virtual void update_deriv();

    virtual SGVector<float64_t> get_derivative_wrt_inference_method(
            const TParameter* param);

    virtual SGVector<float64_t> get_derivative_wrt_likelihood_model(
            const TParameter* param);

    virtual SGVector<float64_t> get_derivative_wrt_kernel(
            const TParameter* param);

    virtual SGVector<float64_t> get_derivative_wrt_mean(
            const TParameter* param);

    virtual SGMatrix<float64_t> get_chol_inv(SGMatrix<float64_t> mtx);

    virtual SGVector<float64_t> compute_mvmK(SGVector<float64_t> al);

    virtual SGVector<float64_t> compute_mvmZ(SGVector<float64_t> x);

    virtual SGVector<float64_t> get_derivative_wrt_inducing_features(const TParameter* param);

    virtual SGVector<float64_t> get_derivative_wrt_inducing_noise(const TParameter* param);

    virtual SGVector<float64_t> derivative_helper_when_Wneg(SGVector<float64_t> res, const TParameter* param);

    virtual float64_t get_derivative_related_cov(SGVector<float64_t> ddiagKi,
        SGMatrix<float64_t> dKuui, SGMatrix<float64_t> dKui);

    virtual float64_t get_derivative_related_mean(SGVector<float64_t> dmu);

    virtual float64_t get_derivative_implicit_term_helper(SGVector<float64_t> d);
private:
    void init();

protected:
    SGVector<float64_t> m_sW;

    SGVector<float64_t> m_d2lp;

    SGVector<float64_t> m_d3lp;

    SGVector<float64_t> m_dlp;

    SGVector<float64_t> m_W;

    float64_t m_tolerance;

    index_t m_iter;

    float64_t m_opt_tolerance;

    float64_t m_opt_max;

    SGMatrix<float64_t> m_chol_R0;

    SGVector<float64_t> m_dfhat;

    SGVector<float64_t> m_g;

    float64_t m_Psi;

    SGVector<float64_t> m_dg;

    bool m_Wneg;
};
}
#endif /* HAVE_EIGEN3 */
#endif /* CSINGLEFITCLAPLACIANINFERENCEMETHOD_H */