Decentralised Machine Learning (DML) enables collaborative machine learning without centralised input data. Federated Learning (FL) and Edge Inference are examples of DML. While tools for DML (especially FL) are starting to flourish, many are not flexible and portable enough to experiment with novel processors (e.g., RISC-V), non-fully connected network topologies, and asynchronous collaboration schemes. We overcome these limitations via a domain-specific language allowing us to map DML schemes to an underlying middleware, i.e. the FastFlow parallel programming library. We experiment with it by generating different working DML schemes on x86-64 and ARM platforms and an emerging RISC-V one. We characterise the performance and energy efficiency of the presented schemes and systems. As a byproduct, we introduce a RISC-V porting of the PyTorch framework, the first publicly available to our knowledge.
This paper presents a study of causality in a reversible, concurrent setting. There exist various notions of causality in π-calculus, which differ in the treatment of parallel extrusions of the same name. In this paper we present a uniform framework for reversible π-calculi that is parametric with respect to a data structure that stores information about an extrusion of a name. Different data structures yield different approaches to the parallel extrusion problem. We map three well-known causal semantics into our framework. We show that the (parametric) reversibility induced by our framework is causallyconsistent and prove a causal correspondence between an appropriate instance of the framework and Boreale and Sangiorgi's causal semantics.