diff --git a/smac/runhistory/runhistory2epm.py b/smac/runhistory/runhistory2epm.py
index 598f7f72f..05e23e17d 100644
--- a/smac/runhistory/runhistory2epm.py
+++ b/smac/runhistory/runhistory2epm.py
@@ -3,6 +3,7 @@
 import typing
 
 import numpy as np
+import scipy as sp
 
 from smac.tae.execute_ta_run import StatusType
 from smac.runhistory.runhistory import RunHistory, RunKey, RunValue
@@ -10,13 +11,14 @@
 from smac.epm.base_imputor import BaseImputor
 from smac.utils import constants
 from smac.scenario.scenario import Scenario
+from smac.utils.constants import VERY_SMALL_NUMBER
 
-__author__ = "Katharina Eggensperger"
-__copyright__ = "Copyright 2015, ML4AAD"
+__author__ = "Katharina Eggensperger, Marius Lindauer"
+__copyright__ = "Copyright 2015-2020, AutoML.org"
 __license__ = "3-clause BSD"
 __maintainer__ = "Katharina Eggensperger"
 __email__ = "eggenspk@cs.uni-freiburg.de"
-__version__ = "0.0.1"
+__version__ = "0.0.3"
 
 
 class AbstractRunHistory2EPM(object):
@@ -545,6 +547,57 @@ def transform_response_values(self, values: np.ndarray) -> np.ndarray:
         return values
 
 
+class RunHistory2EPM4BiLogCost(RunHistory2EPM4Cost):
+        """TODO"""
+
+        def transform_response_values(self, values: np.ndarray) -> np.ndarray:
+            """Transform function response values.
+
+            Transform the response values by using a bilog transformation
+
+            Source: "Scalable Constrained Bayesian Optimization" by Eriksson and Poloczek
+            https://arxiv.org/pdf/2002.08526.pdf
+
+            Parameters
+            ----------
+            values : np.ndarray
+                Response values to be transformed.
+
+            Returns
+            -------
+            np.ndarray
+            """
+            return np.sign(values) * np.log(1 + np.abs(values))
+
+
+class RunHistory2EPM4GaussianCopula(RunHistory2EPM4Cost):
+    """TODO"""
+
+    def transform_response_values(self, values: np.ndarray) -> np.ndarray:
+        """Transform function response values.
+
+        Transform the response values by using a Gaussian Copula:
+            1. compute quantiles of ECDF on given values
+            2. compute inverse Gaussian CDF for these quantiles
+
+        Source: "Scalable Constrained Bayesian Optimization" by Eriksson and Poloczek
+        https://arxiv.org/pdf/2002.08526.pdf
+
+        Parameters
+        ----------
+        values : np.ndarray
+            Response values to be transformed.
+
+        Returns
+        -------
+        np.ndarray
+        """
+        # ECDF
+        quants = [sp.stats.percentileofscore(values, v)/100 - VERY_SMALL_NUMBER for v in values]
+        # Inverse Gaussian CDF
+        return np.array([sp.stats.norm.ppf(q) for q in quants]).reshape((-1, 1))
+
+
 class RunHistory2EPM4EIPS(AbstractRunHistory2EPM):
     """TODO"""