In this paper, we derive a PAC-Bayes bound on the generalisation gap, in a supervised time-series setting for a special class of discrete-time non-linear dynamical systems. This class includes stable recurrent neural networks (RNN), and the motivation for this work was its application to RNNs. In order to achieve the results, we impose some stability constraints, on the allowed models. Here, stability is understood in the sense of dynamical systems. For RNNs, these stability conditions can be expressed in terms of conditions on the weights. We assume the processes involved are essentially bounded and the loss functions are Lipschitz. The proposed bound on the generalisation gap depends on the mixing coefficient of the data distribution, and the essential supremum of the data. Furthermore, the bound converges to zero as the dataset size increases. In this paper, we 1) formalize the learning problem, 2) derive a PAC-Bayesian error bound for such systems, 3) discuss various consequences of this error bound, and 4) show an illustrative example, with discussions on computing the proposed bound. Unlike other available bounds the derived bound holds for non i.i.d. data (time-series) and it does not grow with the number of steps of the RNN.

In this paper we derive a PAC-Bayesian error bound for a class of stochastic dynamical systems with inputs, namely, for linear time-invariant stochastic state-space models (stochastic LTI systems for short). This class of systems is widely used in control engineering and econometrics, in particular, they represent a special case of recurrent neural networks. In this paper we 1) formalize the learning problem for stochastic LTI systems with inputs, 2) derive a PAC-Bayesian error bound for such systems, and 3) discuss various consequences of this error bound.

In this work, we examine Asymmetric Shapley Values (ASV), a variant of the popular SHAP additive local explanation method. ASV proposes a way to improve model explanations incorporating known causal relations between variables, and is also considered as a way to test for unfair discrimination in model predictions. Unexplored in previous literature, relaxing symmetry in Shapley values can have counter-intuitive consequences for model explanation. To better understand the method, we first show how local contributions correspond to global contributions of variance reduction. Using variance, we demonstrate multiple cases where ASV yields counter-intuitive attributions, arguably producing incorrect results for root-cause analysis. Second, we identify generalized additive models (GAM) as a restricted class for which ASV exhibits desirable properties. We support our arguments by proving multiple theoretical results about the method. Finally, we demonstrate the use of asymmetric attributions on multiple real-world datasets, comparing the results with and without restricted model families using gradient boosting and deep learning models.

In this paper, we improve the PAC-Bayesian error bound for linear regression derived in Germain et al. (2016). The improvements are two-fold. First, the proposed error bound is tighter, and converges to the generalization loss with a well-chosen temperature parameter. Second, the error bound also holds for training data that are not independently sampled. In particular, the error bound applies to certain time series generated by well-known classes of dynamical models, such as ARX models.

In this paper, we present a case-study on application of Ramadge-Wonham supervisory control theory (abbreviated by SCT in the sequel) to a patient support system of a magnetic resonance imaging (MRI) scanner. We discuss the whole developmental cycle, starting from the mathematical models of the uncontrolled system and of the control requirements, and ending with the implementation of the obtained controller on the actual hardware. The obtained controller was tested on the physical system. In this case study, we attempted to build the models in a modular way, in order to decrease the computational complexity of the controller synthesis and to improve the adaptability of the models. An important advantage of SCT is that it allows automatic generation of the controller, and that it can thus improve adaptability of the control software. We also briefly discuss our experience on the adaptability of the control software, obtained in the course of this case study.

In this paper we study the identifiability of linear switched systems (LSSs) in discrete-time.The question of identifiability is central to system identification, as it sets the boundaries of applicability of any system identification method; no system identification algorithm can properly estimate the parameters of a system which is not identifiable. We present necessary and sufficient conditions that guarantee structural identifiability for parametrized LSSs. We also introduce the class of semi-algebraic parametrizations, for which these conditions can be checked effectively.