cvaps.h

强化学习算法（R-Learning）难得的珍贵资料

// Copyright (C) 2003
// Gerhard Neumann ([e-mail])

//                
// This file is part of RL Toolbox.
// http://www.igi.tugraz.at/ril_toolbox
//
// 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.
// 3. The name of the author may not be used to endorse or promote products
//    derived from this software without specific prior written permission.
// 
// THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 AUTHOR 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.

#ifndef C__VAPS_H
#define C__VAPS_H

#include "cagentlistener.h"
#include "cstate.h"
#include "cfeaturefunction.h"
#include "cpolicies.h"
#include "cresiduals.h"
#include "cerrorfunction.h"
#include "cpolicygradient.h"
#include "ril_debug.h"

class CGradientEpisodeLearner : public CSemiMDPRewardListener
{
protected:
	CFeatureList *dWTrace;
	CFeatureList *dWBuf;

	int nSteps;

	CGradientVETraces *dWETraces;

	CErrorFunction *errorFunction;
public:
	
	CGradientEpisodeLearner(CRewardFunction *reward, CErrorFunction *error);
	~CGradientEpisodeLearner();


	virtual void nextStep(CStateCollection *oldState, CAction *action, rlt_real reward, CStateCollection *newState);
	virtual void newEpisode();

	virtual void updateGradient();
	virtual void resetTraces();

	void setAlpha(rlt_real value);
	rlt_real getAlpha();

	void setMaxSteps(int maxSteps);
	int getMaxSteps();

	rlt_real getDWETraceFactor() {return getParameter("ETraceFactor");};
	void setDWETraceFactor(rlt_real factor)  {setParameter("ETraceFactor", factor);};
};

class CVAPSGradientLearner : public CGradientEpisodeLearner
{
protected:
	CFeatureList *dTTrace;
	CFeatureList *dTBuf;

	CStochasticPolicy *policy;

public:
	CVAPSGradientLearner(CRewardFunction *reward, CErrorFunction *error, CStochasticPolicy *policy);
	~CVAPSGradientLearner();


	virtual void nextStep(CStateCollection *oldState, CAction *action, rlt_real reward, CStateCollection *newState);

	virtual void resetTraces();
};

class CVAPSGradientCalculator : public CPolicyGradientCalculator, public CVAPSGradientLearner
{
protected:
	CAgent *agent; 

	CFeatureList *gradient;
public:
	CVAPSGradientCalculator(CRewardFunction *reward, CErrorFunction *error, CStochasticPolicy *policy, CAgent *agent, int TSteps);
	~CVAPSGradientCalculator();

	virtual void updateGradient();

	virtual void getGradient(CFeatureList *gradient);
};

#endif