Search

We study robust and efficient distributed algorithms for building and maintaining distributed data structures in dynamic Peer-to-Peer (P2P) networks. P2P networks are characterized by a high level of dynamicity with abrupt heavy node \emph{churn} (nodes that join and leave the network continuously over time). We present a novel algorithm that builds and maintains with high probability a skip list for $poly(n)$ rounds despite $\mathcal{O}(n/\log n)$ churn \emph{per round} ($n$ is the stable network size). We assume that the churn is controlled by an oblivious adversary (that has complete knowledge and control of what nodes join and leave and at what time and has unlimited computational power, but is oblivious to the random choices made by the algorithm). Moreover, the maintenance overhead is proportional to the churn rate. Furthermore, the algorithm is scalable in that the messages are small (i.e., at most $polylog(n)$ bits) and every node sends and receives at most $polylog(n)$ messages per round. Our algorithm crucially relies on novel distributed and parallel algorithms to merge two $n$-elements skip lists and delete a large subset of items, both in $\mathcal{O}(\log n)$ rounds with high probability. These procedures may be of independent interest due to their elegance and potential applicability in other contexts in distributed data structures. To the best of our knowledge, our work provides the first-known fully-distributed data structure that provably works under highly dynamic settings (i.e., high churn rate). Furthermore, they are localized (i.e., do not require any global topological knowledge). Finally, we believe that our framework can be generalized to other distributed and dynamic data structures including graphs, potentially leading to stable distributed computation despite heavy churn.

The industry increasingly relies on deep learning (DL) technology for manufacturing inspections, which are challenging to automate with rule-based machine vision algorithms. DL-powered inspection systems derive defect patterns from labeled images, combining human-like agility with the consistency of a computerized system. However, finite labeled datasets often fail to encompass all natural variations necessitating Continuous Training (CT) to regularly adjust their models with recent data. Effective CT requires fresh labeled samples from the original distribution; otherwise, selfgenerated labels can lead to silent performance degradation. To mitigate this risk, we develop a robust CT-based maintenance approach that updates DL models using reliable data selections through a two-stage filtering process. The initial stage filters out low-confidence predictions, as the model inherently discredits them. The second stage uses variational auto-encoders and histograms to generate image embeddings that capture latent and pixel characteristics, then rejects the inputs of substantially shifted embeddings as drifted data with erroneous overconfidence. Then, a fine-tuning of the original DL model is executed on the filtered inputs while validating on a mixture of recent production and original datasets. This strategy mitigates catastrophic forgetting and ensures the model adapts effectively to new operational conditions. Evaluations on industrial inspection systems for popsicle stick prints and glass bottles using critical real-world datasets showed less than 9% of erroneous self-labeled data are retained after filtering and used for fine-tuning, improving model performance on production data by up to 14% without compromising its results on original validation data.

Optimal cooling that minimises thermal gradients and the average temperature is essential for enhanced battery safety and health. This work presents a new modelling approach for battery cells of different shapes by integrating Chebyshev spectral-Galerkin method and model component decomposition. As a result, a library of reduced-order computationally efficient battery thermal models is obtained, characterised by different numbers of states. These models are validated against a high-fidelity finite element model and are compared with a thermal equivalent circuit (TEC) model under real-world vehicle driving and battery cooling scenarios. Illustrative results demonstrate that the proposed model with four states can faithfully capture the two-dimensional thermal dynamics, while the model with only one state significantly outperforms the widely-used two-state TEC model in both accuracy and computational efficiency, reducing computation time by 28.7\%. Furthermore, our developed models allow for independent control of tab and surface cooling channels, enabling effective thermal performance optimisation. Additionally, the proposed model's versatility and effectiveness are demonstrated through various applications, including the evaluation of different cooling scenarios, closed-loop temperature control, and cell design optimisation., Comment: 13 pages

Electroencephalography (EEG) recordings are analyzed using battery-powered wearable devices to monitor brain activities and neurological disorders. These applications require long and continuous processing to generate feasible results. However, wearable devices are constrained with limited energy and computation resources, owing to their small sizes for practical use cases. Embedded heterogeneous multi-core platforms (HMPs) can provide better performance within limited energy budgets for EEG applications. Error resilience of the EEG application pipeline can be exploited further to maximize the performance and energy gains with HMPs. However, disciplined tuning of approximation on embedded HMPs requires a thorough exploration of the accuracy-performance-power trade-off space. In this work, we characterize the error resilience of three EEG applications, including Epileptic Seizure Detection, Sleep Stage Classification, and Stress Detection on the real-world embedded HMP test-bed of the Odroid XU3 platform. We present a combinatorial evaluation of power-performance-accuracy trade-offs of EEG applications at different approximation, power, and performance levels to provide insights into the disciplined tuning of approximation in EEG applications on embedded platforms., Comment: 7 pages, 10 figures

This paper presents an innovative approach to recognizing personality traits using deep learning (DL) methods applied to electrocardiogram (ECG) signals. Within the framework of detecting the big five personality traits model encompassing extra-version, neuroticism, agreeableness, conscientiousness, and openness, the research explores the potential of ECG-derived spectrograms as informative features. Optimal window sizes for spectrogram generation are determined, and a convolutional neural network (CNN), specifically Resnet-18, and visual transformer (ViT) are employed for feature extraction and personality trait classification. The study utilizes the publicly available ASCERTAIN dataset, which comprises various physiological signals, including ECG recordings, collected from 58 participants during the presentation of video stimuli categorized by valence and arousal levels. The outcomes of this study demonstrate noteworthy performance in personality trait classification, consistently achieving F1-scores exceeding 0.9 across different window sizes and personality traits. These results emphasize the viability of ECG signal spectrograms as a valuable modality for personality trait recognition, with Resnet-18 exhibiting effectiveness in discerning distinct personality traits.

Recommending routes by their probability of having a rider has long been the goal of conventional route recommendation systems. While this maximizes the platform-specific criteria of efficiency, it results in sub-optimal outcomes with the disparity among the income of drivers who work for similar time frames. Pioneer studies on fairness in ridesharing platforms have focused on algorithms that match drivers and riders. However, these studies do not consider the time schedules of different riders sharing a ride in the ridesharing mode. To overcome this shortcoming, we present the first route recommendation system for ridesharing networks that explicitly considers fairness as an evaluation criterion. In particular, we design a routing mechanism that reduces the inequality among drivers and provides them with routes that have a similar probability of finding riders over a period of time. However, while optimizing fairness the efficiency of the platform should not be affected as both of these goals are important for the long-term sustainability of the system. In order to jointly optimize fairness and efficiency we consider repositioning drivers with low income to the areas that have a higher probability of finding riders in future. While applying driver repositioning, we design a future-aware policy and allocate the areas to the drivers considering the destination of requests in the corresponding area. Extensive simulations on real-world datasets of Washington DC and New York demonstrate superior performance by our proposed system in comparison to the existing baselines.