The limits of qualitative simulation

2016

Mechanisms play an important role in many sciences when it comes to questions concerning explanation, prediction, and control. Answering such questions in a quantitative way requires a formal represention of mechanisms. Gebharter (2014) suggests to represent mechanisms by means of one or more causal arrows of an acyclic causal net. In this paper we show how this approach can be extended in such a way that it can also be fruitfully applied to mechanisms featuring causal feedback. Citation information: Gebharter, A., & Schurz, G. (2016). A modelling approach for mechanisms featuring causal cycles. Philosophy of Science, 83(5), 934–945. doi:10.1086/687876

Synthese, forthcoming

Mechanistic philosophy of science views a large part of scientific activity as engaged in modelling mechanisms. While science textbooks tend to offer qualitative models of mechanisms, there is increasing demand for models from which one can draw quantitative predictions and explanations. Casini et al. put forward the Recursive Bayesian Net (RBN) formalism as well suited to this end. The RBN formalism is an extension of the standard Bayesian net formalism, an extension that allows for modelling the hierarchical nature of mechanisms. Like the standard Bayesian net formalism, it models causal relationships using directed acyclic graphs. Given this appeal to acyclicity, causal cycles pose a prima facie problem for the RBN approach. This paper argues that the problem is a significant one given the ubiquity of causal cycles in mechanisms, but that the problem can be solved by combining two sorts of solution strategy in a judicious way.

1996

In this paper we present an event-based approach to qualitative simulation. We suggest that the behaviour of a system with time is best measured in terms of the landmark events that occur i.e. events that result in interesting changes to the system being modelled. For us, a behaviour model corresponds not to a sequence of qualitative state descriptions but to a set of event sequences | the things that actually happen to the system rather than the way it happens to be at certain times. Although we have a simple implementation of our system, our primary purpose in developing it is to derive a high level, event-based, nonmonotonic language for specifying qualitative simulation systems. We not only illustrate how a qualitative simulation program can be directly speci ed (and implemented) in our language, we also sketch how qualitative simulation systems from the literature can be de ned and reconstructed in our calculus.

Knowledge Science, Engineering and Management, 2007

Many chemistry students have difficulty in understanding an organic chemistry subject called reaction mechanisms. Mastering the subject would require the application of chemical intuition and chemical commonsense adequately. This work discusses a novel framework using Qualitative Reasoning (QR) to provide means for learning reaction mechanisms through simulation. The framework consists of a number of functional components. These include substrate recognizer, qualitative model constructor, prediction engine, molecule update routine, explanation generator, and a knowledge base containing essential chemical facts and chemical theories. Chemical processes are represented as qualitative models using Qualitative Process Theory (QPT) ontology. The construction of these models is automated based on a set of QR algorithms. We have tested the framework on the S N 1 and the S N 2 reaction mechanisms. Representative cases of reaction simulation and causal explanation are also included to demonstrate how these models can serve as a cognitive tool fostering the acquisition of conceptual understanding via qualitative simulation.

Qualitative simulation is a well-known reasoning technique that involves the use of simulation technologies. Reasoning is made to determine qualitative values and change directions of system variables, and it is done for each time point and time interval following the time point. Qualitative variables possess continuous qualitative value sets that are discretized by landmark points. Qualitative simulation uses qualitative time representation and its quantitative value is of no interest. The main purpose of this study was to develop a technique to determine time steps for a quantitative simulation under guidance of qualitative information. The proposed technique determined time advances using qualitative and quantitative information together to obtain a robust time step as wide as possible for simulation time advances. For this purpose, sign algebraic properties and derivation roots of quantitative equations and qualitative variable values with their change directions were used to compute time advances. In the approach, qualitative simulation determined landmark points to be advanced, and quantitative simulation calculated the duration required. Using the proposed algorithm, the simulation is advanced instead of iterating simulation time for a predefined time step and checking whether or not there is any activity in the interval, directly to the time points that are qualitatively different.

2018

The study of actual causation concerns reasoning about events that have been instrumental in bringing about a particular outcome. Although the subject has long been studied in a number of fields including artificial intelligence, existing approaches have not yet reached the point where their results can be directly applied to explain causation in certain advanced scenarios, such as pin-pointing causes and responsibilities for the behavior of a complex cyber-physical system. We believe that this is due, at least in part, to a lack of distinction between the laws that govern individual states of the world and events whose occurrence cause state to evolve. In this paper, we present a novel approach to reasoning about actual causation that leverages techniques from Reasoning about Actions and Change to identify detailed causal explanations for how an outcome of interest came to be. We also present an implementation of the approach that leverages Answer Set Programming.

2010

This paper presents a divide-and-conquer approach that aims at making QSIM simulation tractable. We consider dynamical systems the structure of which can be represented by compartments. The system model is decomposed into sub-models tightly connected through shared variables on the basis of the analysis of the causality relations intrinsically captured by the compartmental model structure. The sub-models are separately simulated but their behaviors are constrained by the information on the shared variables generated from the simulation. The partition of the complete model into smaller ones prevents the construction of temporal correlations between variables in different sub-models, and thus the generation of a complete temporal ordering of all unrelated events that is one of the major causes of intractable branching in qualitative simulation. The strategy we propose is discussed through a case study in the eld of Plant Pathology, namely the germination process of Plasmopara viticola...

Artificial Intelligence, 1990

This paper examines qualitative simulation (QS) from the phase space perspective of dynamic systems theory. QS consists of two steps: transition analysis determines the sequence of qualitative states that a system traverses and global interpretation derives its long-term behavior . I recast transition analysis as a search problem in phase space and replace the assorted transition rules with two algebraic conditions. The first condition determines transitions between arbitrarily shaped regions in phase space, as opposed to QS which only handles n-dimensional rectangles . It also provides more accurate results by considering only the boundaries between regions . The second condition determines whether nearby trajectories approach a fixed point asymptotically. It obtains better results than QS by exploiting local stability properties . I recast global interpretation as a search for attractors in phase space and present a global interpretation algorithm for systems whose local behavior determines global behavior uniquely.

BMC Systems Biology, 2016

Background: Qualitative dynamics semantics provide a coarse-grain modeling of networks dynamics by abstracting away kinetic parameters. They allow to capture general features of systems dynamics, such as attractors or reachability properties, for which scalable analyses exist. The Systems Biology Graphical Notation Process Description language (SBGN-PD) has become a standard to represent reaction networks. However, no qualitative dynamics semantics taking into account all the main features available in SBGN-PD had been proposed so far. Results: We propose two qualitative dynamics semantics for SBGN-PD reaction networks, namely the general semantics and the stories semantics, that we formalize using asynchronous automata networks. While the general semantics extends standard Boolean semantics of reaction networks by taking into account all the main features of SBGN-PD, the stories semantics allows to model several molecules of a network by a unique variable. The obtained qualitative models can be checked against dynamical properties and therefore validated with respect to biological knowledge. We apply our framework to reason on the qualitative dynamics of a large network (more than 200 nodes) modeling the regulation of the cell cycle by RB/E2F. Conclusion: The proposed semantics provide a direct formalization of SBGN-PD networks in dynamical qualitative models that can be further analyzed using standard tools for discrete models. The dynamics in stories semantics have a lower dimension than the general one and prune multiple behaviors (which can be considered as spurious) by enforcing the mutual exclusiveness between the activity of different nodes of a same story. Overall, the qualitative semantics for SBGN-PD allow to capture efficiently important dynamical features of reaction network models and can be exploited to further refine them.

Polarity and causality are important concepts but have not received much attention in the system dynamics literature. The great effort it takes students to properly understand them has motivated this inquiry. In the framework of a conceptual model of interacting with complex systems, several cognitive tasks are proposed. This paper concentrates on one of them that deals with causal links' polarity. An examination of other approaches that deal with causality and use more or less similar diagram languages shows that usually causality is only very broadly defined, and where it is operationally defined, this is done with respect to events rather than behavior. In contrast to these approaches, system dynamics is about behavior rather than events. We then revisit the traditional criticism of causal loop diagrams and show a way out, but add two new criticisms related to the inability of causal loop diagrams to address behavior: in fact it seems that they are closer to the event-related definition of causality. Also, the impossibility to execute them in simulations means that executable concept-models are to be preferred: they express important information a causal loop diagram cannot represent and on top of it they render the behavioral consequences visible (as opposed to the events). In conclusion, causal loop diagrams should only be used by experienced modelers, and be banned from educational use.