Multivariate Time Series (MTS) data capture temporal behaviors to provide invaluable insights into various physical dynamic phenomena. In smart mobility, MTS plays a crucial role in providing temporal dynamics of behaviors such as maneuver patterns, enabling early detection of anomalous behaviors while facilitating pro-activity in Prognostics and Health Management (PHM). In this work, we aim to address challenges associated with modeling MTS data collected from a vehicle using sensors. Our goal is to investigate the effectiveness of two distinct unsupervised representation learning approaches in identifying maneuvering states in smart mobility. Specifically, we focus on some bivariate accelerations extracted from 2.5 years of driving, where the dataset is non-stationary, long, noisy, and completely unlabeled, making manual labeling impractical. The approaches of interest are Temporal Neighborhood Coding for Maneuvering (TNC4Maneuvering) and Decoupled Local and Global Representation learner for Maneuvering (DLG4Maneuvering). The main advantage of these frameworks is that they capture transferable insights in a form of representations from the data that can be effectively applied in multiple subsequent tasks, such as time-series classification, clustering, and multi-linear regression, which are the quantitative measures and qualitative measures, including visualization of representations themselves and resulting reconstructed MTS, respectively. We compare their effectiveness, where possible, in order to gain insights into which approach is more effective in identifying maneuvering states in smart mobility.

Federated Learning (FL) is an innovative approach to distributed machine learning. While FL offers significant privacy advantages, it also faces security challenges, particularly from poisoning attacks where adversaries deliberately manipulate local model updates to degrade model performance or introduce hidden backdoors. Existing defenses against these attacks have been shown to be effective when the data on the nodes is identically and independently distributed (i.i.d.), but they often fail under less restrictive, non-i.i.d data conditions. To overcome these limitations, we introduce Celtibero, a novel defense mechanism that integrates layered aggregation to enhance robustness against adversarial manipulation. Through extensive experiments on the MNIST and IMDB datasets, we demonstrate that Celtibero consistently achieves high main task accuracy (MTA) while maintaining minimal attack success rates (ASR) across a range of untargeted and targeted poisoning attacks. Our results highlight the superiority of Celtibero over existing defenses such as FL-Defender, LFighter, and FLAME, establishing it as a highly effective solution for securing federated learning systems against sophisticated poisoning attacks.

Increasingly large imitation learning datasets are being collected with the goal of training foundation models for robotics. However, despite the fact that data selection has been of utmost importance in vision and natural language processing, little work in robotics has questioned what data such models should actually be trained on. In this work we investigate how to weigh different subsets or ``domains'' of robotics datasets for robot foundation model pre-training. Concrete, we use distributionally robust optimization (DRO) to maximize worst-case performance across all possible downstream domains. Our method, Re-Mix, addresses the wide range of challenges that arise when applying DRO to robotics datasets including variability in action spaces and dynamics across different datasets. Re-Mix employs early stopping, action normalization, and discretization to counteract these issues. Through extensive experimentation on the largest open-source robot manipulation dataset, the Open X-Embodiment dataset, we demonstrate that data curation can have an outsized impact on downstream performance. Specifically, domain weights learned by Re-Mix outperform uniform weights by 38\% on average and outperform human-selected weights by 32\% on datasets used to train existing generalist robot policies, specifically the RT-X models.

Stock price prediction is a challenging problem in the field of finance and receives widespread attention. In recent years, with the rapid development of technologies such as deep learning and graph neural networks, more research methods have begun to focus on exploring the interrelationships between stocks. However, existing methods mostly focus on the short-term dynamic relationships of stocks and directly integrating relationship information with temporal information. They often overlook the complex nonlinear dynamic characteristics and potential higher-order interaction relationships among stocks in the stock market. Therefore, we propose a stock price trend prediction model named LSR-IGRU in this paper, which is based on long short-term stock relationships and an improved GRU input. Firstly, we construct a long short-term relationship matrix between stocks, where secondary industry information is employed for the first time to capture long-term relationships of stocks, and overnight price information is utilized to establish short-term relationships. Next, we improve the inputs of the GRU model at each step, enabling the model to more effectively integrate temporal information and long short-term relationship information, thereby significantly improving the accuracy of predicting stock trend changes. Finally, through extensive experiments on multiple datasets from stock markets in China and the United States, we validate the superiority of the proposed LSR-IGRU model over the current state-of-the-art baseline models. We also apply the proposed model to the algorithmic trading system of a financial company, achieving significantly higher cumulative portfolio returns compared to other baseline methods. Our sources are released at https://github.com/ZP1481616577/Baselines\_LSR-IGRU.

ESG Rating Disagreement and Corporate Total Factor Productivity:Inference and Prediction Zhanli Li ESG rating disagreement can lead to a decline in corporate total factor productivity When faced with ESG rating disagreement, reactive green innovation by enterprises does not lead to improvements in total factor productivity. Abstract This paper explores the relationship between ESG rating disagreement and total factor productivity (TFP) based on data from Chinese domestic ESG rating agencies and financial data of A-share listed companies in China from 2015 to 2022. On one hand, the empirical results show that ESG rating disagreement reduces corporate TFP, a conclusion that is validated through multiple robustness tests. The mechanism analysis reveals an interaction effect between green innovation and ESG rating disagreement. Specifically, in firms without ESG rating disagreement, green innovation promotes the improvement of TFP; however, in firms with disagreement, although ESG rating disagreement may drive green innovation, this does not lead to an increase in TFP. The heterogeneity analysis indicates that this effect is more pronounced in non-state-owned, asset-intensive, and lowpollution enterprises.

Developmental dysgraphia is a neurological disorder that hinders children's writing skills. In recent years, researchers have increasingly explored machine learning methods to support the diagnosis of dysgraphia based on offline and online handwriting. In most previous studies, the two types of handwriting have been analysed separately, which does not necessarily lead to promising results. In this way, the relationship between online and offline data cannot be explored. To address this limitation, we propose a novel multimodal machine learning approach utilizing both online and offline handwriting data. We created a new dataset by transforming an existing online handwritten dataset, generating corresponding offline handwriting images. We considered only different types of word data (simple word, pseudoword & difficult word) in our multimodal analysis. We trained SVM and XGBoost classifiers separately on online and offline features as well as implemented multimodal feature fusion and soft-voted ensemble. Furthermore, we proposed a novel ensemble with conditional feature fusion method which intelligently combines predictions from online and offline classifiers, selectively incorporating feature fusion when confidence scores fall below a threshold. Our novel approach achieves an accuracy of 88.8%, outperforming SVMs for single modalities by 12-14%, existing methods by 8-9%, and traditional multimodal approaches (soft-vote ensemble and feature fusion) by 3% and 5%, respectively. Our methodology contributes to the development of accurate and efficient dysgraphia diagnosis tools, requiring only a single instance of multimodal word/pseudoword data to determine the handwriting impairment. This work highlights the potential of multimodal learning in enhancing dysgraphia diagnosis, paving the way for accessible and practical diagnostic tools.

In online continual learning (CL), models trained on changing distributions easily forget previously learned knowledge and bias toward newly received tasks. To address this issue, we present Continual Bias Adaptor (CBA), a bi-level framework that augments the classification network to adapt to catastrophic distribution shifts during training, enabling the network to achieve a stable consolidation of all seen tasks. However, the CBA module adjusts distribution shifts in a class-specific manner, exacerbating the stability gap issue and, to some extent, fails to meet the need for continual testing in online CL. To mitigate this challenge, we further propose a novel class-agnostic CBA module that separately aggregates the posterior probabilities of classes from new and old tasks, and applies a stable adjustment to the resulting posterior probabilities. We combine the two kinds of CBA modules into a unified Dual-CBA module, which thus is capable of adapting to catastrophic distribution shifts and simultaneously meets the real-time testing requirements of online CL. Besides, we propose Incremental Batch Normalization (IBN), a tailored BN module to re-estimate its population statistics for alleviating the feature bias arising from the inner loop optimization problem of our bi-level framework. To validate the effectiveness of the proposed method, we theoretically provide some insights into how it mitigates catastrophic distribution shifts, and empirically demonstrate its superiority through extensive experiments based on four rehearsal-based baselines and three public continual learning benchmarks.

Continuous glucose monitoring (CGM) devices provide real-time glucose monitoring and timely alerts for glycemic excursions, improving glycemic control among patients with diabetes. However, identifying rare events like hypoglycemia and hyperglycemia remain challenging due to their infrequency. Moreover, limited access to sensitive patient data hampers the development of robust machine learning models. Our objective is to accurately predict glycemic excursions while addressing data privacy concerns. To tackle excursion prediction, we propose a novel Hypo-Hyper (HH) loss function, which significantly improves performance in the glycemic excursion regions. The HH loss function demonstrates a 46% improvement over mean-squared error (MSE) loss across 125 patients. To address privacy concerns, we propose FedGlu, a machine learning model trained in a federated learning (FL) framework. FL allows collaborative learning without sharing sensitive data by training models locally and sharing only model parameters across other patients. FedGlu achieves a 35% superior glycemic excursion detection rate compared to local models. This improvement translates to enhanced performance in predicting both, hypoglycemia and hyperglycemia, for 105 out of 125 patients. These results underscore the effectiveness of the proposed HH loss function in augmenting the predictive capabilities of glucose predictions. Moreover, implementing models within a federated learning framework not only ensures better predictive capabilities but also safeguards sensitive data concurrently.

Simulation of vehicle motion in planetary environments is challenging. This is due to the modeling of complex terrain, optical conditions, and terrain-aware vehicle dynamics. One of the critical issues of typical simulators is that they assume terrain is a rigid body, which limits their ability to render wheel traces and compute the wheel-terrain interactions. This prevents, for example, the use of wheel traces as landmarks for localization, as well as the accurate simulation of motion. In the context of lunar regolith, the surface is not rigid but granular. As such, there are differences in the rover's motion, such as sinkage and slippage, and a clear wheel trace left behind the rover, compared to that on a rigid terrain. This study presents a novel approach to integrating a terramechanics-aware terrain deformation engine to simulate a realistic wheel trace in a digital lunar environment. By leveraging Discrete Element Method simulation results alongside experimental single-wheel test data, we construct a regression model to derive deformation height as a function of contact normal force. The region of interest in a height map is retrieved from the wheel poses. The elevation values of corresponding pixels are subsequently modified using contact normal forces and the regression model. Finally, we apply the determined elevation change to each mesh vertex to render wheel traces during runtime. The deformation engine is integrated into our ongoing development of a lunar simulator based on NVIDIA's Omniverse IsaacSim. We hypothesize that our work will be crucial to testing perception and downstream navigation systems under conditions similar to outdoor or terrestrial fields. A demonstration video is available here: https://www.youtube.com/watch?v=TpzD0h-5hv4

Simulation is a common approach to predict the effect of business process changes on quantitative performance. The starting point of Business Process Simulation (BPS) is a process model enriched with simulation parameters. To cope with the typically large parameter spaces of BPS models, several methods have been proposed to automatically discover BPS models from event logs. Virtually all these approaches neglect the data perspective of business processes. Yet, the data attributes manipulated by a business process often determine which activities are performed, how many times, and when. This paper addresses this gap by introducing a data-aware BPS modeling approach and a method to discover data-aware BPS models from event logs. The BPS modeling approach supports three types of data attributes (global, case-level, and event-level) as well as deterministic and stochastic attribute update rules and data-aware branching conditions. An empirical evaluation shows that the proposed method accurately discovers the type of each data attribute and its associated update rules, and that the resulting BPS models more closely replicate the process execution control flow relative to data-unaware BPS models.