protected double getBoundLP(double countLog, ref RlStats stats)
{
var value = getValueEstimate(ref stats);
//if (m_searchParams.biasTermConstant == 0.0) return value;
value += m_searchParams.biasTermConstant * Math.Sqrt(countLog / (stats.m_count + 1));
return(value);
}
protected double getValueEstimate(ref RlStats stats)
{
return(stats.m_count > 0 ? stats.m_mean : m_firstPlayUrgency);
}
public RlAction(Action_t action)
{
m_stats = new RlStats();
m_action = action;
}
