Actionable Interpretability through Optimizable Counterfactual Explanations for Tree Ensembles.

Ana Lucic; Harrie Oosterhuis; Hinda Haned; Maarten de Rijke

Actionable Interpretability through Optimizable Counterfactual Explanations for Tree Ensembles.

Ana Lucic
, Harrie Oosterhuis
, Hinda Haned
, Maarten de Rijke

Research output: Working paper › Preprint

Original language	American English
State	Published - Nov 27 2019

Disciplines

Computer Sciences
Artificial Intelligence and Robotics

Cite this

@techreport{f570e5ebe50048da8140e8a14f39425c,

title = "Actionable Interpretability through Optimizable Counterfactual Explanations for Tree Ensembles.",

author = "Ana Lucic and Harrie Oosterhuis and Hinda Haned and \{de Rijke\}, Maarten",

note = "Model interpretability has become an important problem in ML due to the increased effect algorithmic decisions have on humans. Counterfactual explanations can help users understand not only why ML models make certain decisions, but also how these decisions can be changed.",

year = "2019",

month = nov,

day = "27",

language = "American English",

type = "WorkingPaper",

}

TY - UNPB

T1 - Actionable Interpretability through Optimizable Counterfactual Explanations for Tree Ensembles.

AU - Lucic, Ana

AU - Oosterhuis, Harrie

AU - Haned, Hinda

AU - de Rijke, Maarten

N1 - Model interpretability has become an important problem in ML due to the increased effect algorithmic decisions have on humans. Counterfactual explanations can help users understand not only why ML models make certain decisions, but also how these decisions can be changed.

PY - 2019/11/27

Y1 - 2019/11/27

UR - https://arxiv.org/abs/1911.12199v2

M3 - Preprint

BT - Actionable Interpretability through Optimizable Counterfactual Explanations for Tree Ensembles.

ER -

Actionable Interpretability through Optimizable Counterfactual Explanations for Tree Ensembles.

Disciplines

Other files and links

Cite this