Abstract We propose a new sequential decision-making setting, combining key aspects of two established online learning problems with bandit feedback. The optimal action to play at any given moment is contingent on an underlying changing state that is not directly observable by the agent. Each state is associated with a context distribution, possibly corrupted, allowing the agent to identify the state. Furthermore, states evolve in a Markovian fashion, providing useful information to estimate the current state via state history. In the proposed problem setting, we tackle the challenge of deciding on which of the two sources of information the agent should base its action selection. We present an algorithm that uses a referee to dynamically combine the policies of a contextual bandit and a multi-armed bandit. We capture the time-correlation of states through iteratively learning the action-reward transition model, allowing for efficient exploration of actions. Our setting is motivated by adaptive mobile health (mHealth) interventions. Users transition through different, time-correlated, but only partially observable internal states, determining their current needs. The side information associated with each internal state might not always be reliable, and standard approaches solely rely on the context risk of incurring high regret. Similarly, some users might exhibit weaker correlations between subsequent states, leading to approaches that solely rely on state transitions risking the same. We analyze our setting and algorithm in terms of regret lower bound and upper bounds and evaluate our method on simulated medication adherence intervention data and several real-world data sets, showing improved empirical performance compared to several popular algorithms.

ABSTRACT This paper proposes a two‐stage framework that combines uncertainty reduction and predictive modelling to enhance the accuracy of battery Remaining Useful Life (RUL) prediction. In the first stage, a simplified fuzzy optimization learning model is introduced to mitigate uncertainty caused by abnormal capacity fluctuations in battery data. The proposed fuzzy model reconstructs degradation data into a consistent downward trend based on mid‐ and short‐term tendencies of the battery, alleviating abnormal variability and improving suitability for predictive modelling. In the second stage, uncertainty arising during the recursive prediction process of a standalone Transformer model was mitigated through the integration of a particle filter. This approach dynamically manages prediction errors using particles, effectively controlling cumulative errors and enhancing the stability and reliability of long‐term predictions. This methodology can lead to extended battery life and increased operational reliability through accurate RUL prediction. The proposed methodology is validated through experiments using NASA and CALCE battery datasets, demonstrating superior prediction accuracy and stability compared to conventional approaches by systematically reducing uncertainties.

Graph neural networks (GNNs), especially dynamic GNNs, have become a research hotspot in spatiotemporal forecasting problems. While many dynamic graph construction methods have been developed, relatively few of them explore the causal relationship between neighbor nodes. Thus, the resulting models lack strong explainability for the causal relationship between the neighbor nodes of the dynamically generated graphs, which can easily lead to a risk in subsequent decisions. Moreover, few of them consider the uncertainty and noise of dynamic graphs based on the time series datasets, which are ubiquitous in real-world graph structure networks. In this article, we propose a novel dynamic diffusion-variational GNN (DVGNN) for spatiotemporal forecasting. For dynamic graph construction, an unsupervised generative model is devised. Two layers of graph convolutional network (GCN) are applied to calculate the posterior distribution of the latent node embeddings in the encoder stage. Then, a diffusion model is used to infer the dynamic link probability and reconstruct causal graphs (CGs) in the decoder stage adaptively. The new loss function is derived theoretically, and the reparameterization trick is adopted in estimating the probability distribution of the dynamic graphs by evidence lower bound (ELBO) during the backpropagation period. After obtaining the generated graphs, dynamic GCN and temporal attention are applied to predict future states. Experiments are conducted on four real-world datasets of different graph structures in different domains. The results demonstrate that the proposed DVGNN model outperforms state-of-the-art approaches and achieves outstanding root mean square error (RMSE) results while exhibiting higher robustness. Also, by F1-score and probability distribution analysis, we demonstrate that DVGNN better reflects the causal relationship and uncertainty of dynamic graphs. The website of the code is https://github.com/gorgen2020/DVGNN.

Exploring the missing values is an essential but challenging issue due to the complex latent spatio-temporal correlation and dynamic nature of time series. Owing to the outstanding performance in dealing with structure learning potentials, Graph Neural Networks (GNNs) and Recurrent Neural Networks (RNNs) are often used to capture such complex spatio-temporal features in multivariate time series. However, these data-driven models often fail to capture the essential spatio-temporal relationships when significant signal corruption occurs. Additionally, calculating the high-order neighbor nodes in these models is of high computational complexity. To address these problems, we propose a novel higher-order spatio-temporal physics-incorporated GNN (HSPGNN). Firstly, the dynamic Laplacian matrix can be obtained by the spatial attention mechanism. Then, the generic inhomogeneous partial differential equation (PDE) of physical dynamic systems is used to construct the dynamic higher-order spatio-temporal GNN adaptively to obtain the missing time series values. Moreover, we estimate the missing impact by Normalizing Flows (NF) to evaluate the importance of each node in the graph for better explainability. Experimental results on four benchmark datasets demonstrate the effectiveness of HSPGNN and the superior performance when combining various order neighbor nodes. Also, graph-like optical flow, dynamic graphs, and missing impact can be obtained naturally by HSPGNN, which provides better dynamic analysis and explanation than traditional data-driven models.

In the rapidly evolving field of AI, Explainable Artificial Intelligence (XAI) has become paramount, particularly in Intelligent Environments applications. It offers clarity and understanding in complex decision-making processes, fostering trust and enabling rigorous scrutiny. The Shapley value, renowned for its accurate quantification of feature importance, has emerged as a prevalent standard in both academic research and practical application. Nevertheless, the Shapley value’s reliance on the calculation of all possible coalitions poses a significant computational challenge, as it falls within the class of NP-hard problems. Consequently, approximation techniques are employed in most practical scenarios as a substitute for precise computations. The most common of those is the SHAP (SHapley Additive exPlanations) technique, which quantifies the influence exerted by a specific feature on decision outcomes of a specific Machine Learning model. However, the Shapley value’s theoretical underpinnings focus on assessing and understanding feature impact on model evaluation metrics, rather than just alterations in the responses. This paper conducts a comparative analysis using controlled synthetic data with established ground truths. It juxtaposes the practical implementation of the SHAP approach with the theoretical model in two distinct scenarios: one using the F1-score and the other, the accuracy metric. These are two representative characteristic functions, capturing different aspects and whose appropriateness depends on the specific requirements and context of the task to be solved. We analyze how the three alternatives exhibit similarity and disparity in their manifestation of feature effects. We explore the parallels and differences between these approaches in reflecting feature effects. Ultimately, our research seeks to determine the conditions under which SHAP outcomes are more aligned with either the F1-score or the accuracy metric, thereby providing valuable insights for their application in various Intelligent Environment contexts.

This paper presents a comprehensive empirical investigation into the interactions between various randomization techniques in Deep Neural Networks (DNNs) and their impact on learning performance. It is well-established that injecting randomness into the training process of DNNs, through various approaches, at different stages, is often beneficial for reducing overfitting and improving generalization. Nonetheless, the interactions between randomness techniques such as weight noise, dropout, and many others remain poorly understood. Consequently, it is challenging to determine which methods can be effectively combined to optimize DNN performance. To address this issue, we categorize the existing randomness techniques into four key types: injection of noise/randomness at the data, model structure, optimization or learning stage. We use this classification to identify gaps in the current coverage of potential mechanisms for the introduction of randomness, leading to proposing two new techniques: adding noise to the loss function and random masking of the gradient updates. In our empirical study, we employ a Particle Swarm Optimizer (PSO) for hyperparameter optimization (HPO) to explore the space of possible configurations to determine where and how much randomness should be injected to maximize DNN performance. We assess the impact of various types and levels of randomness for DNN architectures across standard computer vision benchmarks: MNIST, FASHION-MNIST, CIFAR10, and CIFAR100. Across more than 30 000 evaluated configurations, we perform a detailed examination of the interactions between randomness techniques and their combined impact on DNN performance. Our findings reveal that randomness through data augmentation and in weight initialization are the main contributors to performance improvement. Additionally, correlation analysis demonstrates that different optimizers, such as Adam and Gradient Descent with Momentum, prefer distinct types of randomization during the training process. A GitHub repository with the complete implementation and generated dataset is available1.

The field of Explainable Predictive Maintenance (PM) is concerned with developing methods that can clarify how AI systems operate in the PM domain. One of the challenges of creating maintenance plans is integrating AI output with human decision-making pro- cesses and expertise. For AI to be helpful and trustworthy, fault predictions must be contextualized and easily comprehensible to humans. This involves providing tailored explanations to different actors depending on their roles and needs. For example, engineers can be connected to technical installation blueprints, while man- agers can evaluate system downtime costs, and lawyers can assess safety-threatening failures' potential liability. In many industries, black-box AI systems analyze sensor data to predict failures by detecting anomalies and deviations from typical behavior with impressive accuracy. However, PM is just one part of a broader context that aims to identify the most probable causes, develop a recovery plan, and estimate remaining useful life while providing alternative solutions. Achieving this requires complex interactions among various actors in industrial and decision-making processes. Our tutorial explores current trends, and promising research directions in Explainable AI (XAI) relevant to Explainable Predictive Maintenance (XPM), and future challenges and open issues on this topic. We will also present three case studies that highlight XPM's challenges in bus and train operations and steel factories.

As Deep Neural Networks (DNN) make their way into safety-critical decision processes, it becomes imperative to have robust and reliable uncertainty estimates for their predictions for both in-distribution and out-of-distribution (OOD) examples. This is particularly important in real-life high-risk settings such as healthcare, where OOD examples (e.g., patients with previously unseen or rare labels, i.e., diagnoses) are frequent, and an incorrect clinical decision might put human life in danger, in addition to having severe ethical and financial costs. While evidential uncertainty estimates for deep learning have been studied for multi-class problems, research in multi-label settings remains untapped. In this paper, we propose a DEep Evidential Doctor (DEED), which is a novel deterministic approach to estimate multi-label targets along with uncertainty. We achieve this by placing evidential priors over the original likelihood functions and directly estimating the parameters of the evidential distribution using a novel loss function. Additionally, we build a redundancy layer (particularly for high uncertainty and OOD examples) to minimize the risk associated with erroneous decisions based on dubious predictions. We achieve this by learning the mapping between the evidential space and a continuous semantic label embedding space via a recurrent decoder. Thereby inferring, even in the case of OOD examples, reasonably close predictions to avoid catastrophic consequences. We demonstrate the effectiveness of DEED on a digit classification task based on a modified multi-label MNIST dataset, and further evaluate it on a diagnosis prediction task from a real-life electronic health record dataset. We highlight that in terms of prediction scores, our approach is on par with the existing state-of-the-art having a clear advantage of generating reliable, memory and time-efficient uncertainty estimates with minimal changes to any multi-label DNN classifier.

Research on machine activity recognition (MAR) is drawing more attention because MAR can provide productivity monitoring for efficiency optimization, better maintenance scheduling, product design improvement, and potential material savings. A particular challenge of MAR for human-operated machines is the overlap when transiting from one activity to another: during transitions, operators often perform two activities simultaneously, e.g., lifting the fork already while approaching a rack, so the exact time when one activity ends and another begins is uncertain. Machine learning models are often uncertain during such activity transitions, and we propose a novel ensemble-based method adapted to fuzzy transitions in a forklift MAR problem. Unlike traditional ensembles, where models in the ensemble are trained on different subsets of data, or with costs that force them to be diverse in their responses, our approach is to train a single model that predicts several activity labels, each under a different context. These individual predictions are not made by independent networks but are made using a structure that allows for sharing important features, i.e., a context ensemble. The results show that the gated recurrent unit network can provide medium or strong confident context ensembles for 95% of the cases in the test set, and the final forklift MAR result achieves accuracies of 97% for driving and 90% for load-handling activities. This study is the first to highlight the overlapping activity issue in MAR problems and to demonstrate that the recognition results can be significantly improved by designing a machine learning framework that addresses this issue.

Evolutionary Algorithms (EAs) are often challenging to apply in real-world settings since evolutionary computations involve a large number of evaluations of a typically expensive fitness function. For example, an evaluation could involve training a new machine learning model. An approximation (also known as meta-model or a surrogate) of the true function can be used in such applications to alleviate the computation cost. In this paper, we propose a two-stage surrogate-assisted evolutionary approach to address the computational issues arising from using Genetic Algorithm (GA) for feature selection in a wrapper setting for large datasets. We define “Approximation Usefulness” to capture the necessary conditions to ensure correctness of the EA computations when an approximation is used. Based on this definition, we propose a procedure to construct a lightweight qualitative meta-model by the active selection of data instances. We then use a meta-model to carry out the feature selection task. We apply this procedure to the GA-based algorithm CHC (Cross generational elitist selection, Heterogeneous recombination and Cataclysmic mutation) to create a Qualitative approXimations variant, CHC Q X . We show that CHC Q X converges faster to feature subset solutions of significantly higher accuracy (as compared to CHC), particularly for large datasets with over 100K instances. We also demonstrate the applicability of the thinking behind our approach more broadly to Swarm Intelligence (SI), another branch of the Evolutionary Computation (EC) paradigm with results of PSO Q X , a qualitative approximation adaptation of the Particle Swarm Optimization (PSO) method. A GitHub repository with the complete implementation is available. 2

Domain adaptation (DA) methods facilitate cross-domain learning by minimizing the marginal or conditional distribution shift between domains. However, the conditional distribution shift is not well addressed by existing DA techniques for the cross-domain regression learning task. In this paper, we propose Multi-Domain Adaptation for Regression under Conditional shift (DARC) method. DARC constructs a shared feature space such that linear regression on top of that space generalizes to all domains. In other words, DARC aligns different domains according to the task-related information encoded in the values of the dependent variable. It is achieved using a novel Pairwise Similarity Preserver (PSP) loss function. PSP incentivizes the differences between the outcomes of any two samples, regardless of their domain(s), to match the distance between these samples in the constructed space. We perform experiments in both two-domain and multi-domain settings. The two-domain setting is helpful, especially when one domain contains few available labeled samples and can benefit from adaptation to a domain with many labeled samples. The multi-domain setting allows several domains, each with limited data, to be adapted collectively; thus, multiple domains compensate for each other’s lack of data. The results from all the experiments conducted both on synthetic and real-world datasets confirm the effectiveness of DARC.

In the latent bandit problem, the learner has access to reward distributions and – for the non-stationary variant – transition models of the environment. The reward distributions are conditioned on the arm and unknown latent states. The goal is to use the reward history to identify the latent state, allowing for the optimal choice of arms in the future. The latent bandit setting lends itself to many practical applications, such as recommender and decision support systems, where rich data allows the offline estimation of environment models with online learning remaining a critical component. Previous solutions in this setting always choose the highest reward arm according to the agent’s beliefs about the state, not explicitly considering the value of information-gathering arms. Such information-gathering arms do not necessarily provide the highest reward, thus may never be chosen by an agent that chooses the highest reward arms at all times. In this paper, we present a method for information-gathering in latent bandits. Given particular reward structures and transition matrices, we show that choosing the best arm given the agent’s beliefs about the states incurs higher regret. Furthermore, we show that by choosing arms carefully, we obtain an improved estimation of the state distribution, and thus lower the cumulative regret through better arm choices in the future. Through theoretical analysis we show that the proposed method retains the sub-linear regret rate of previous methods while having much better problem dependent constants. We evaluate our method on both synthetic and real-world data sets, showing significant improvement in regret over state-of-the-art methods.

Using hierarchies of classes is one of the standard methods to solve multi-class classification problems. In the literature, selecting the right hierarchy is considered to play a key role in improving classification performance. Although different methods have been proposed, there is still a lack of understanding of what makes a hierarchy good and what makes a method to extract hierarchies perform better or worse. To this effect, we analyze and compare some of the most popular approaches to extracting hierarchies. We identify some common pitfalls that may lead practitioners to make misleading conclusions about their methods. To address some of these problems, we demonstrate that using random hierarchies is an appropriate benchmark to assess how the hierarchy’s quality affects the classification performance. In particular, we show how the hierarchy’s quality can become irrelevant depending on the experimental setup: when using powerful enough classifiers, the final performance is not affected by the quality of the hierarchy. We also show how comparing the effect of the hierarchies against non-hierarchical approaches might incorrectly indicate their superiority. Our results confirm that datasets with a high number of classes generally present complex structures in how these classes relate to each other. In these datasets, the right hierarchy can dramatically improve classification performance.

Mobility as a Service (MaaS) is a relatively new domain where new types of knowledge systems have recently emerged.It combines various modes of transportation and different kinds of data to present personalized services to travellers based on transport needs.A knowledge-based framework based on Artificial Intelligence (AI) is proposed in this paper to integrate, analyze, and process different types of mobility data.The framework includes a knowledge acquisition process to extract and structure data from various sources, including mobility experts and add new information to a knowledge base.The role of AI in this framework is to aid in automatically discovering knowledge from various data sets and recommend efficient and personalized mobility services with explanations.A scenario is also presented to demonstrate the interaction of the proposed framework's modules.

Abstract In contextual anomaly detection, an object is only considered anomalous within a specific context. Most existing methods use a single context based on a set of user-specified contextual features. However, identifying the right context can be very challenging in practice, especially in datasets with a large number of attributes. Furthermore, in real-world systems, there might be multiple anomalies that occur in different contexts and, therefore, require a combination of several “useful” contexts to unveil them. In this work, we propose a novel approach, called wisdom of the contexts (WisCon), to effectively detect complex contextual anomalies in situations where the true contextual and behavioral attributes are unknown. Our method constructs an ensemble of multiple contexts, with varying importance scores, based on the assumption that not all useful contexts are equally so. We estimate the importance of each context using an active learning approach with a novel query strategy. Experiments show that WisCon significantly outperforms existing baselines in different categories (i.e., active learning methods, unsupervised contextual and non-contextual anomaly detectors) on 18 datasets. Furthermore, the results support our initial hypothesis that there is no single perfect context that successfully uncovers all kinds of contextual anomalies, and leveraging the “wisdom” of multiple contexts is necessary.

Feature selection is an intractable problem, therefore practical algorithms often trade off the solution accuracy against the computation time. In this paper, we propose a novel multi-stage feature selection framework utilizing multiple levels of approximations, or surrogates. Such a framework allows for using wrapper approaches in a much more computationally efficient way, significantly increasing the quality of feature selection solutions achievable, especially on large datasets. We design and evaluate a Surrogate-Assisted Genetic Algorithm (SAGA) which utilizes this concept to guide the evolutionary search during the early phase of exploration. SAGA only switches to evaluating the original function at the final exploitation phase.We prove that the run-time upper bound of SAGA surrogate-assisted stage is at worse equal to the wrapper GA, and it scales better for induction algorithms of high order of complexity in number of instances. We demonstrate, using 14 datasets from the UCI ML repository, that in practice SAGA significantly reduces the computation time compared to a baseline wrapper Genetic Algorithm (GA), while converging to solutions of significantly higher accuracy. Our experiments show that SAGA can arrive at near-optimal solutions three times faster than a wrapper GA, on average. We also showcase the importance of evolution control approach designed to prevent surrogates from misleading the evolutionary search towards false optima.

Batteries are a safety-critical and the most expensive component for electric vehicles (EVs). To ensure the reliability of the EVs in operation, it is crucial to monitor the state of health of those batteries. Monitoring their deterioration is also relevant to the sustainability of the transport solutions, through creating an efficient strategy for utilizing the remaining capacity of the battery and its second life. Electric buses, similar to other EVs, come in many different variants, including different configurations and operating conditions. Developing new degradation models for each existing combination of settings can become challenging from different perspectives such as unavailability of failure data for novel settings, heterogeneity in data, low amount of data available for less popular configurations, and lack of sufficient engineering knowledge. Therefore, being able to automatically transfer a machine learning model to new settings is crucial. More concretely, the aim of this work is to extract features that are invariant across different settings. In this study, we propose an evolutionary method, called genetic algorithm for domain invariant features (GADIF), that selects a set of features to be used for training machine learning models, in such a way as to maximize the invariance across different settings. A Genetic Algorithm, with each chromosome being a binary vector signaling selection of features, is equipped with a specific fitness function encompassing both the task performance and domain shift. We contrast the performance, in migrating to unseen domains, of our method against a number of classical feature selection methods without any transfer learning mechanism. Moreover, in the experimental result section, we analyze how different features are selected under different settings. The results show that using invariant features leads to a better generalization of the machine learning models to an unseen domain.

Ensemble learning methods combine multiple models to improve performance by exploiting their diversity. The success of these approaches relies heavily on the dissimilarity of the base models forming the ensemble. This diversity can be achieved in many ways, with well-known examples including bagging and boosting. It is the diversity of the models within an ensemble that allows the ensemble to correct the errors made by its members, and consequently leads to higher classification or regression performance. A mistake made by a base model can only be rectified if other members behave differently on that particular instance, and provide the aggregator with enough information to make an informed decision. On the contrary, lack of diversity not only lowers model performance, but also wastes computational resources. Nevertheless, in the current state of the art ensemble approaches, there is no guarantee on the level of diversity achieved, and no mechanism ensuring that each member will learn a different decision boundary from the others. In this paper, we propose a parallel orthogonal deep learning architecture in which diversity is enforced by design, through imposing an orthogonality constraint. Multiple deep neural networks are created, parallel to each other. At each parallel layer, the outputs of different base models are subject to Gram-Schmidt orthogonalization. We demonstrate that this approach leads to a high level of diversity from two perspectives. First, the models make different errors on different parts of feature space, and second, they exhibit different levels of uncertainty in their decisions. Experimental results confirm the benefits of the proposed method, compared to standard deep learning models and well-known ensemble methods, in terms of diversity and, as a result, classification performance.

Unscheduled 30-day readmissions are a hallmark of Congestive Heart Failure (CHF) patients that pose significant health risks and escalate care cost. In order to reduce readmissions and curb the cost of care, it is important to initiate targeted intervention programs for patients at risk of readmission. This requires identifying high-risk patients at the time of discharge from hospital. Here, using real data from over 7500 CHF patients hospitalized between 2012 and 2016 in Sweden, we built and tested a deep learning framework to predict 30-day unscheduled readmission. We present a cost-sensitive formulation of Long Short-Term Memory (LSTM) neural network using expert features and contextual embedding of clinical concepts. This study targets key elements of an Electronic Health Record (EHR) driven prediction model in a single framework: using both expert and machine derived features, incorporating sequential patterns and addressing the class imbalance problem. We evaluate the contribution of each element towards prediction performance (ROC-AUC, F1-measure) and cost-savings. We show that the model with all key elements achieves higher discrimination ability (AUC: 0.77; F1: 0.51; Cost: 22% of maximum possible savings) outperforming the reduced models in at least two evaluation metrics. Additionally, we present a simple financial analysis to estimate annual savings if targeted interventions are offered to high risk patients.

In the era of big data, considerable research focus is being put on designing efficient algorithms capable of learning and extracting high-level knowledge from ubiquitous data streams in an online fashion. While, most existing algorithms assume that data samples are drawn from a stationary distribution, several complex environments deal with data streams that are subject to change over time. Taking this aspect into consideration is an important step towards building truly aware and intelligent systems. In this paper, we propose GNG-A, an adaptive method for incremental unsupervised learning from evolving data streams experiencing various types of change. The proposed method maintains a continuously updated network (graph) of neurons by extending the Growing Neural Gas algorithm with three complementary mechanisms, allowing it to closely track both gradual and sudden changes in the data distribution. First, an adaptation mechanism handles local changes where the distribution is only non-stationary in some regions of the feature space. Second, an adaptive forgetting mechanism identifies and removes neurons that become irrelevant due to the evolving nature of the stream. Finally, a probabilistic evolution mechanism creates new neurons when there is a need to represent data in new regions of the feature space. The proposed method is demonstrated for anomaly and novelty detection in non-stationary environments. Results show that the method handles different data distributions and efficiently reacts to various types of change.

An approach for intelligent monitoring of mobile cyberphysical systems is described, based on consensus among distributed self-organised agents. Its usefulness is experimentally demonstrated over a long-time case study in an example domain: a fleet of city buses. The proposed solution combines several techniques, allowing for life-long learning under computational and communication constraints. The presented work is a step towards autonomous knowledge discovery in a domain where data volumes are increasing, the complexity of systems is growing, and dedicating human experts to build fault detection and diagnostic models for all possible faults is not economically viable. The embedded, self-organised agents operate on-board the cyberphysical systems, modelling their states and communicating them wirelessly to a back-office application. Those models are subsequently compared against each other to find systems which deviate from the consensus. In this way the group (e.g., a fleet of vehicles) is used to provide a standard, or to describe normal behaviour, together with its expected variability under particular operating conditions. The intention is to detect faults without the need for human experts to anticipate them beforehand. This can be used to build up a knowledge base that accumulates over the life-time of the systems. The approach is demonstrated using data collected during regular operation of a city bus fleet over the period of almost 4 years.

With an increasing amount of data in intelligent transportation systems, methods are needed to automatically extract general representations that accurately predict not only known tasks but also similar tasks that can emerge in the future. Creation of low-dimensional representations can be unsupervised or can exploit various labels in multi-task learning (when goal tasks are known) or transfer learning (when they are not) settings. Finding a general, low-dimensional representation suitable for multiple tasks is an important step toward knowledge discovery in aware intelligent transportation systems. This paper evaluates several approaches mapping high-dimensional sensor data from Volvo trucks into a low-dimensional representation that is useful for prediction. Original data are bivariate histograms, with two types-turbocharger and engine-considered. Low-dimensional representations were evaluated in a supervised fashion by mean equal error rate (EER) using a random forest classifier on a set of 27 1-vs-Rest detection tasks. Results from unsupervised learning experiments indicate that using an autoencoder to create an intermediate representation, followed by t -distributed stochastic neighbor embedding, is the most effective way to create low-dimensional representation of the original bivariate histogram. Individually, t -distributed stochastic neighbor embedding offered best results for 2-D or 3-D and classical autoencoder for 6-D or 10-D representations. Using multi-task learning, combining unsupervised and supervised objectives on all 27 available tasks, resulted in 10-D representations with a significantly lower EER compared to the original 400-D data. In transfer learning setting, with topmost diverse tasks used for representation learning, 10-D representations achieved EER comparable to the original representation

Fuel used by heavy duty trucks is a major cost for logistics companies, and therefore improvements in this area are highly desired. Many of the factors that influence fuel consumption, such as the road type, vehicle configuration or external environment, are difficult to influence. One of the most under-explored ways to lower the costs is training and incentivizing drivers. However, today it is difficult to measure driver performance in a comprehensive way outside of controlled, experimental setting. This paper proposes a machine learning methodology for quantifying and qualifying driver performance, with respect to fuel consumption, that is suitable for naturalistic driving situations. The approach is a knowledge-based feature extraction technique, constructing a normalizing fuel consumption value denoted Fuel under Predefined Conditions (FPC), which captures the effect of factors that are relevant but are not measured directly. The FPC, together with information available from truck sensors, is then compared against the actual fuel used on a given road segment, quantifying the effects associated with driver behavior or other variables of interest. We show that raw fuel consumption is a biased measure of driver performance, being heavily influenced by other factors such as high load or adversary weather conditions, and that using FPC leads to more accurate results. In this paper we also show evaluation the proposed method using large-scale, real-world, naturalistic database of heavy-duty vehicle operation.

We study agents situated in partially observable environments, who do not have sufficient resources to create conformant plans. Instead, they generate plans which are conditional and partial, execute or simulate them, and learn to evaluate their quality from experience. Our agent employs an incomplete symbolic deduction system based on Active Logic and Situation Calculus for reasoning about actions and their consequences. An Inductive Logic Programming algorithm generalises observations and deduced knowledge, allowing the agent to execute a good plan. We show results of using PROGOL learning algorithm to distinguish "bad" plans early in the reasoning process, before too many resources are wasted on considering them. We show that additional knowledge needs to be provided before learning can be successful, but argue that the benefits achieved make it worthwhile. Finally, we identify several assumptions made by PROGOL, shared by other similarly universal algorithms, which are well justified in general, but fail to exploit the properties of the class of problems faced by rational agents.

The paper is an experimental study of using the rough sets based rule induction algorithm MODLEM in the framework of multiple classifiers. Particular attention is paid to using a meta-classifier called combiner, which learns how to aggregate answers of component classifiers. The experimental results confirm that the range of classification improvement for the combiner depends on the independence of errors made by the component classifiers. Moreover, we summarize the experience with using MODLEM in other multiple classifiers, namely the bagging and n/sup 2/ classifiers.