Juliana Bowles

Mohit Kumar, Alexander Valentinitsch, Magdalena Fuchs et al.

This paper develops a novel mathematical framework for collaborative learning by means of geometrically inspired kernel machines which includes statements on the bounds of generalisation and approximation errors, and sample complexity. For classification problems, this approach allows us to learn bounded geometric structures around given data points and hence solve the global model learning problem in an efficient way by exploiting convexity properties of the related optimisation problem in a Reproducing Kernel Hilbert Space (RKHS). In this way, we can reduce classification problems to determining the closest bounded geometric structure from a given data point. Further advantages that come with our solution is that our approach does not require clients to perform multiple epochs of local optimisation using stochastic gradient descent, nor require rounds of communication between client/server for optimising the global model. We highlight that numerous experiments have shown that the proposed method is a competitive alternative to the state-of-the-art.

Juliana Bowles, Marco B. Caminati

Complex systems are usually modelled through a combination of structural and behavioural models, where separate behavioural models make it easier to design and understand partial behaviour. When partial models are combined, we need to guarantee that they are consistent, and several automated techniques have been developed to check this. We argue that in some cases it is impossible to guarantee total consistency, and instead we want to find execution paths across such models with minimal conflicts with respect to a certain metric of interest. We present an efficient and scalable solution to find optimal paths through a combination of the theorem prover Isabelle with the constraint solver Z3. Our approach has been inspired by a healthcare problem, namely how to detect conflicts between medications taken by patients with multiple chronic conditions, and how to find preferable alternatives automatically.