ACCBench: A Framework for Comparing Causality Algorithms

  title={ACCBench: A Framework for Comparing Causality Algorithms},
  author={Simon Rehwald and Amjad Ibrahim and Kristian Beckers and Alexander Pretschner},
Modern socio-technical systems are increasingly complex. A fundamental problem is that the borders of such systems are often not well-defined a-priori, which among other problems can lead to unwanted behavior during runtime. Ideally, unwanted behavior should be prevented. If this is not possible the system shall at least be able to help determine potential cause(s) a-posterori, identify responsible parties and make them accountable for their behavior. Recently, several algorithms… 

Figures and Tables from this paper

Actual Causality Canvas: A General Framework for Explanation-Based Socio-Technical Constructs

This paper proposes a unifying framework and an interactive platform (Actual Causality Canvas) to address the problem of operationalizing actual causality for different domains and purposes and shows that with minimal effort,actual causality reasoning can be integrated into these domains.

Understanding and Formalizing Accountability for Cyber-Physical Systems

This formal model can be used to highlight context specific-elements of accountability mechanisms, define their capabilities, and express different notions of accountability, and makes design decisions explicit and facilitates discussion, analysis and comparison of different approaches.

2 3 O ct 2 01 8 Understanding and Formalizing Accountability for Cyber-Physical Systems

Accountability is the property of a system that enables the uncovering of causes for events and helps understand who or what is responsible for these events. Definitions and interpretations of



Towards a Unified Model of Accountability Infrastructures

This paper motivates and proposes a model for accountability infrastructures that is expressive enough to capture all of these domains and shows how this model can be instantiated for several real-world use cases.

A General Trace-Based Framework of Logical Causality

This paper presents an approach to blame components based on a single execution trace violating a safety property, and relies on counterfactual reasoning to distinguish component failures that actually contributed to the outcome from failures that had little or no impact on the violation of \(P\).

Auditing with incomplete logs

An auditing framework is presented to assist analysts in find-ing a valid explanation for the events recorded in the logs and to pinpoint policy violations if such an explanation does not exist, when logs are incomplete and two strategies for the refinement of plausible explanations of con-formity are introduced.

Blaming in component-based real-time systems

This paper formalizes blaming using counterfactual reasoning (“what would have been the outcome if component C had behaved correctly?”) to distinguish component failures that actually contributed to the outcome from failures that had no impact on the violation of P.

Time Series Analysis with the Causality Workbench

What the Causality Workbench platform offers for the analysis of causality in time series analysis is described.

Distinguishing Cause from Effect Using Observational Data: Methods and Benchmarks

Empirical results on real-world data indicate that certain methods are indeed able to distinguish cause from effect using only purely observational data, although more benchmark data would be needed to obtain statistically significant conclusions.

Causes and Explanations: A Structural-Model Approach. Part II: Explanations

New definitions of (causal) explanation are proposed, using structural equations to model counterfactuals and it is shown that the definition handles well a number of problematic examples from the literature.

Causes and Explanations: A Structural-Model Approach. Part I: Causes

We propose a new definition of actual causes, using structural equations to model counterfactuals. We show that the definition yields a plausible and elegant account of causation that handles well

Process mining: Using CPN tools to create test logs for mining algorithms

CP-nets are extended to generate XML event logs that can be mined by process mining tools supporting this format and benefit from the simulation capabilities of CPN Tools and, therefore, avoid reinventing the wheel.

Two Concepts of Accountability: Accountability as a Virtue and as a Mechanism

This paper distinguishes between two main concepts of accountability: accountability as a virtue and accountability as a mechanism. In the former case, accountability is used primarily as a normative