Malcolm Egan

Mohamed Seif, Malcolm Egan, Andrea J. Goldsmith et al.

AI-based sensing at wireless edge devices has the potential to significantly enhance Artificial Intelligence (AI) applications, particularly for vision and perception tasks such as in autonomous driving and environmental monitoring. AI systems rely both on efficient model learning and inference. In the inference phase, features extracted from sensing data are utilized for prediction tasks (e.g., classification or regression). In edge networks, sensors and model servers are often not co-located, which requires communication of features. As sensitive personal data can be reconstructed by an adversary, transformation of the features are required to reduce the risk of privacy violations. While differential privacy mechanisms provide a means of protecting finite datasets, protection of individual features has not been addressed. In this paper, we propose a novel framework for privacy-preserving AI-based sensing, where devices apply transformations of extracted features before transmission to a model server.

Tan-Khiem Huynh, Malcolm Egan, Giovanni Neglia et al.

Federated learning (FL) is now recognized as a key framework for communication-efficient collaborative learning. Most theoretical and empirical studies, however, rely on the assumption that clients have access to pre-collected data sets, with limited investigation into scenarios where clients continuously collect data. In many real-world applications, particularly when data is generated by physical or biological processes, client data streams are often modeled by non-stationary Markov processes. Unlike standard i.i.d. sampling, the performance of FL with Markovian data streams remains poorly understood due to the statistical dependencies between client samples over time. In this paper, we investigate whether FL can still support collaborative learning with Markovian data streams. Specifically, we analyze the performance of Minibatch SGD, Local SGD, and a variant of Local SGD with momentum. We answer affirmatively under standard assumptions and smooth non-convex client objectives: the sample complexity is proportional to the inverse of the number of clients with a communication complexity comparable to the i.i.d. scenario. However, the sample complexity for Markovian data streams remains higher than for i.i.d. sampling.

Jan Mrkos, Jan Drchal, Malcolm Egan et al.

This document serves as a technical report for the analysis of on-demand transport dataset. Moreover we show how the dataset can be used to develop a market formation algorithm based on machine learning. Data used in this work comes from Liftago, a Prague based company which connects taxi drivers and customers through a smartphone app. The dataset is analysed from the machine-learning perspective: we give an overview of features available as well as results of feature ranking. Later we propose the SImple Data-driven MArket Formation (SIDMAF) algorithm which aims to improve a relevance while connecting customers with relevant drivers. We compare the heuristics currently used by Liftago with SIDMAF using two key performance indicators.