Disaster classification and victim detection are two important tasks in enabling efficient rescue operations. In this paper, we propose a multi-task learning (MTL) model which accomplishes these two tasks simultaneously. The idea is to attach one pruned head model to another backbone network. We mathematically pinpoint the optimal branching location and the depth of the pruned head model. Apart from the decoupled task training capability, the MTL model offers lesser memory requirements (12.8 MB saving) and better disaster classification accuracy (1-2% gain), while preserving the same detection performance (0.694 of average precision (AP)), as compared to the traditional method. Such advantages of flexibility, speed and accuracy facilitate the large-scale deployment of Internet of Things (IoT) applications, where we explore the potential of federated learning (FL) and active learning (AL). Given the high ambiguity within disaster images, a modified version of AL-based technique is introduced. For realistic implementation, production-ready OpenFL and OpenVINO tools are adopted to update the global FL model and to optimize the trained model, respectively. Experiment results are promising: the FL-based techniques are comparable to or better than their centralized learning (CL) counterparts. Also, our application portability is demonstrated via different hardware such as CPU and Raspberry Pi.

Enabling personalized interactions and gaining deeper insights into user behavior has become increasingly important for improving human-computer interaction. However, the natural class imbalance in personality datasets, reflecting the real-world distribution of traits where certain personality types are less represented, poses a major challenge for classification models and often results in biased performance. This research explores a range of class imbalance mitigation techniques (CIMTs), including sampling methods and loss functions, to address this issue. We introduce Adaptive Focal Loss with Personality-Stratified Dataset Splitting, a novel approach specifically designed to mitigate class imbalance while stabilizing performance in multi-dimensional personality recognition. Additionally, we analyze multiple evaluation techniques., including regular accuracy, F1 score, and balanced accuracy, recommending the latter for a more comprehensive and fair performance analysis. Our experiments reveal that the proposed stratified techniques with label representation are vital for making the performance not sensitive to dataset splitting, while Adaptive Focal Loss significantly enhances classification performance on imbalanced datasets by incorporating trainable hyperparameter, also effectively addressing the challenges of hyperparameter sensitivity and selection. On average across dimensions, our method improves balance accuracy by up to 7% on the Kaggle dataset and 5% on the Essays dataset compared to regular training, while maintaining minimal computational overhead. These findings mark a critical step toward more robust and equitable personality recognition systems using relatively computationally efficient models. The related research resources, including the code repository and datasets, are available and can be accessed at: https://research.jingjietan.com/?q=AFLPS

This paper presents an Enhanced Particle Swarm Optimisation (EPSO) algorithm to improve multi-robot path planning by integrating a new path planning scheme with a cubic Bezier curve trajectory smoothing algorithm. Traditional PSO algorithms often result in suboptimal paths with numerous turns, necessitating frequent stops and higher energy consumption. The proposed EPSO algorithm addresses these issues by generating smoother paths that reduce the number of turns and enhance the efficiency of multi-robot systems. The proposed algorithm was evaluated through simulations in two scenarios, and its performance was compared against the basic PSO algorithm. The results demonstrated that EPSO consistently produced shorter, smoother paths with fewer directional changes, albeit with slightly longer execution times. This improvement translates to more efficient navigation, reduced energy consumption, and enhanced overall performance of multi-robot systems. The findings underscore the potential of EPSO in applications requiring precise and efficient path planning, highlighting its contribution to advancing the field of robotics.

Federated learning (FL) is a technique that allows multiple clients to collaboratively train a global model without sharing their sensitive and bandwidth-hungry data. This paper presents a joint early client termination and local epoch adjustment for FL. We consider the challenges of heterogeneous Internet of Things (IoT) environments including non-independent and identically distributed (non-IID) data as well as diverse computing and communication capabilities. The goal is to strike the best tradeoff among three conflicting objectives, namely global model accuracy, training latency and communication cost. We first leverage the balanced-MixUp technique to mitigate the influence of non-IID data on the FL convergence rate. A weighted sum optimization problem is then formulated and solved via our proposed FL double deep reinforcement learning (FedDdrl) framework, which outputs a dual action. The former indicates whether a participating FL client is dropped, whereas the latter specifies how long each remaining client needs to complete its local training task. Simulation results show that FedDdrl outperforms the existing FL scheme in terms of overall tradeoff. Specifically, FedDdrl achieves higher model accuracy by about 4% while incurring 30% less latency and communication costs.

Respiratory rate (RR) is a critical vital sign that can provide valuable insights into various medical conditions, including pneumonia. Unfortunately, manual RR counting is often unreliable and discontinuous. Current RR estimation algorithms either lack the necessary accuracy or demand extensive window sizes. In response to these challenges, this study introduces a novel method for continuously estimating RR from photoplethysmogram (PPG) with a reduced window size and lower processing requirements. To evaluate and compare classical and deep learning algorithms, this study leverages the BIDMC and CapnoBase datasets, employing the Respiratory Rate Estimation (RRest) toolbox. The optimal classical techniques combination on the BIDMC datasets achieves a mean absolute error (MAE) of 1.9 breaths/min. Additionally, the developed neural network model utilises convolutional and long short-term memory layers to estimate RR effectively. The best-performing model, with a 50% train–test split and a window size of 7 s, achieves an MAE of 2 breaths/min. Furthermore, compared to other deep learning algorithms with window sizes of 16, 32, and 64 s, this study’s model demonstrates superior performance with a smaller window size. The study suggests that further research into more precise signal processing techniques may enhance RR estimation from PPG signals.