Local deterministic string-to-string transductions are generalizations of morphisms on free monoids. Learning local transductions reduces to inference of monoid morphisms. However, learning a restricted class of morphisms, the so-called fine morphisms, is an intractable problem, because the decision version of the empirical risk minimization problem contains an NP-complete subproblem.

Uniform deviation bounds limit the difference between a model’s expected loss and its loss on an empirical sample uniformly for all models in a learning problem. As such, they are a critical component to empirical risk minimization. In this paper, we provide a novel framework to obtain uniform deviation bounds for loss functions which are unbounded. In our main application, this allows us to ob...

Theoretical relative ages of galactic globular clusters have recently been challenged by a semi-empirical relation. It was used to point out that tested sets of isochrones were unable to reproduce the relation and yield internally inconsistent relative ages. We find that differential cluster ages derived with the isochrones by Salaris & Weiss (1998) are reliable and internally consistent. We al...

The agenda in music research that is broadly recognized as embodied music cognition has arrived hand-in-hand with a social interpretation of music, focusing on the real-world basis of its performance, and fostering an empirical approach to musician movement regarding the communicative function and potential of those movements. However, embodied cognition emerged from traditional cognitivism, wh...

This paper provides a review of empirical research on the link between financial intermediation by banks and economic growth. Special attention is paid to the issues of causality, non-linearity, time perspective, financial intermediation proxies, and interaction terms. The review shows that there are still quite a few unresolved issues in empirical research, which causes scepticism towards prio...

Recent work introduced loss functions which measure the error of a prediction based on multiple simultaneous observations or outcomes. In this paper, we explore the theoretical and practical questions that arise when using such multi-observation losses for regression on data sets of (x, y) pairs. When a loss depends on only one observation, the average empirical loss decomposes by applying the ...

It has been recently shown that, under the margin (or low noise) assumption, there exist classifiers attaining fast rates of convergence of the excess Bayes risk, i.e., the rates faster than n−1/2. The works on this subject suggested the following two conjectures: (i) the best achievable fast rate is of the order n−1, and (ii) the plug-in classifiers generally converge slower than the classifie...

We propose a novel framework for the differentially private ERM, input perturbation. Existing differentially private ERM implicitly assumed that the data contributors submit their private data to a database expecting that the database invokes a differentially private mechanism for publication of the learned model. In input perturbation, each data contributor independently randomizes her/his dat...

Improving the operational efficiency of processes is an important goal of business process management (BPM). There exist many proposals with regard to process improvement patterns (PIPs) as practices that aim at supporting this task. Nevertheless, there is still a gap with respect to validating PIPs in terms of their actual business value for a specific organization. Based on empirical research...