Dr Peipei (Paul) Wu

The European Union's (EU) data strategy aims to create a single market for seamless data flow while ensuring proper governance, privacy, and data protection. In this paper, we present SEDIMARK, an EU project, that builds on this strategy by developing a fully decentralised, secure data marketplace. The goal of SEDIMARK is to build a complete toolbox that enables users to purchase and process data assets. The toolbox includes tools for data cleaning, decentralised machine learning models and secure data exchange. SEDIMARK offers users full control over data assets by enabling them to keep their data locally and thus removing the need for central servers. With customisable pipelines and tools, SEDIMARK supports a wide range of users, from novices to experts, promoting seamless collaboration and fair access to high-quality datasets across Europe. The decentralised connectivity in SEDIMARK is achieved with the use of Distributed Ledger Technology (DLT). Furthermore, SEDIMARK's architecture features a unique Connector component using Self Sovereign Identities (SSI), fostering trust and secure interactions. Transactions in SEDIMARK are stored in a Registry, a decentralised, immutable, non-repudiable and permissionless database. Together the technologies used in SEDIMARK ensure privacy, trust and data quality for secure management, sharing, and monetisation of assets in data spaces.

Deep learning models have achieved state-of-the-art performance across numerous domains, but their increasing size and computational complexity pose significant challenges for deployment in resource-constrained environments. Model pruning is a key technique to address this issue by reducing the number of model parameters. However, existing methods often present a trade-off between compression rate, computational speed-up, and performance preservation. This paper introduces a novel hybrid pruning methodology that strategically combines Weight Statistics Aware Pruning (WSAP)-based unstructured pruning with hardware-friendly structured channel pruning. Our approach first determines WSAP-driven pruning ratios using a heuristic based on the weights' Coefficient of Variation (CoV), allowing for more aggressive pruning of less critical layers. It then applies both fine-grained and channel-based pruning to maximize model compression while preserving accuracy. We demonstrate the effectiveness and generality of our method on two diverse tasks: Video Quality Assessment (VQA) with the DOVER-Mobile model and Time-Series Forecasting with the CrossFormer model. Our results show that the proposed hybrid method achieves a superior balance of efficiency and performance, reducing model parameters by up to 80% and FLOPs by over 50% while maintaining the accuracy of the original models. These improvements make our method well-suited for trustworthy and efficient deployment of deep learning models in shared and constrained environments.

Intensity Particle Flow (IPF) SMC-PHD has been proposed recently for multi-target tracking. In this paper, we extend IPF-SMC-PHD filter to distributed setting, and develop a novel consensus method for fusing the estimates from individual sensors, based on Arithmetic Average (AA) fusion. Different from conventional AA method which may be degraded when unreliable estimates are presented, we develop a novel arithmetic consensus method to fuse estimates from each individual IPF-SMC-PHD filter with partial consensus. The proposed method contains a scheme for evaluating the reliability of the sensor nodes and preventing unreliable sensor information to be used in fusion and communication in sensor network, which help improve fusion accuracy and reduce sensor communication costs. Numerical simulations are performed to demonstrate the advantages of the proposed algorithm over the uncooperative IPF-SMC-PHD and distributed particle-PHD with AA fusion.

Audio-visual tracking of multiple speakers requires to estimate the state (e.g. velocity and location) of each speaker by leveraging the information of both audio and visual modalities. Estimating the number of speakers and their states jointly remains a challenging problem. We propose an Audio-Visual Possion Multi-Bernoulli Mixture Filter (AV-PMBM) that can not only predict the number of speakers but also give accurate estimation of their states. We also propose a novel sound source localization technique based on DOA information and a deep learning based object detector to provide reliable audio measurements for the AV tracker. To our knowledge, this represents the first attempt using PMBM for multi-speaker tracking with audio visual modalities. Experiments on the AV16.3 dataset demonstrate that AV-PMBM achieves state-of-the-art performance in optimal sub-pattern assignment (OSPA).