Automated machine learning (AutoML) streamlines the creation of ML models, but few specialized methods have approached the challenging domain of time series forecasting. Deep neural networks (DNNs) often deliver state-of-the-art predictive performance for forecasting data, however these models are also criticized for being computationally intensive black boxes. As a result, when searching for the "best" model, it is crucial to also acknowledge other aspects, such as interpretability and resource consumption. In this paper, we propose AutoXPCR - a novel method that produces DNNs for forecasting under consideration of multiple objectives in an automated and explainable fashion. Our approach leverages meta-learning to estimate any model's performance along PCR criteria, which encompass (P)redictive error, (C)omplexity, and (R)esource demand. Explainability is addressed on multiple levels, as AutoXPCR pro-vides by-product explanations of recommendations and allows to interactively control the desired PCR criteria importance and trade-offs. We demonstrate the practical feasibility AutoXPCR across 108 forecasting data sets from various domains. Notably, our method outperforms competing AutoML approaches - on average, it only requires 20% of computation costs for recommending highly efficient models with 85% of the empirical best quality.

Several machine learning models have been used to tackle time series forecasting. However, it is generally accepted that none of them is universally valid for every application and over time. Therefore, adequate and adaptive real-time model selection is often required to cope with the time-evolving nature of time series and the fact that models have specific Regions of Competence (RoCs) across the time series. In this paper, we perform an online single model selection for time series forecasting by using an adaptive clustering method to compute the RoCs of candidate models. This method can be extended to ensemble base models selection by combining clustering with a rank-based approach. In this framework, the appropriate model(s) is selected online, and the RoCs responsible for model selection update is done adaptively in an informed manner following concept drift detection in the RoCs’ structure. Moreover, the computed RoCs can be used to provide suitable explanations for the reason for selecting certain model(s) in a certain time interval or instant. Since the RoCs are computed independently of the family of forecasting models in question, the explanations we provide are model-agnostic. An extensive empirical study on various real-world datasets shows that our method achieves excellent or on-par results compared to state-of-the-art approaches and various baselines.

Tree-based models have been successfully applied to a wide variety of tasks, including time series forecasting. They are increasingly in demand and widely accepted because of their comparatively high level of interpretability. However, many of them suffer from the overfitting problem, which limits their application in real-world decision-making. This problem becomes even more severe in online-forecasting settings where time series observations are incrementally acquired, and the distributions from which they are drawn may keep changing over time. In this context, we propose a novel method for the online selection of tree-based models using the TreeSHAP explainability method in the task of time series forecasting. We start with an arbitrary set of different tree-based models. Then, we outline a performance-based ranking with a coherent design to make TreeSHAP able to specialize the tree-based forecasters across different regions in the input time series. In this framework, adequate model selection is performed online, adaptively following drift detection in the time series. In addition, explainability is supported on three levels, namely online input importance, model selection, and model output explanation. An extensive empirical study on various real-world datasets demonstrates that our method achieves excellent or on-par results in comparison to the state-of-the-art approaches as well as several baselines.

Both complex and evolving nature of time series structure make forecasting among one of the most important and challenging tasks in time series analysis. Typical methods for forecasting are designed to model time-evolving dependencies between data observations. However, it is generally accepted that none of them is universally valid for every application. Therefore, methods for learning heterogeneous ensembles by combining a diverse set of forecasts together appear as a promising solution to tackle this task. Several approaches, ranging from simple and enhanced averaging tactics to applying meta-learning methods, have been proposed to learn how to combine individual models in an ensemble. However, finding the optimal strategy for ensemble aggregation remains an open research question, particularly, when the ensemble needs to be adapted in real time. In this paper, we leverage a deep reinforcement learning framework for learning linearly weighted ensembles as a meta-learning method. In this framework, the combination policy in ensembles is modelled as a sequential decision making process, and an actor-critic model aims at learning the optimal weights in a continuous action space. The policy is updated following a drift detection mechanism for tracking performance shifts of the ensemble model. An extensive empirical study on many real-world datasets demonstrates that our method achieves excellent or on par results in comparison to the state-of-the-art approaches as well as several baselines.

Ensemble models are widely used as an effective technique in time-series forecasting, and recently, are inclined toward leveraging meta-learning methods due to their proven predictive advantages in combining individual models in an ensemble. However, finding the optimal strategy for ensemble aggregation is an open research question, particularly, when the ensemble needs to be adapted in real-time. In this paper, we pro-pose a novel meta-learning approach for aggregation of linearly weighted ensembles for the task of time-series forecasting. We outline a deep reinforcement learning framework with a coherent design of the components of the environment and the objective function as an aggregation method in our task. In this framework, the combination policy in ensembles is modeled as a sequential decision making process which is able to capture the temporal behavior in time-series, and an actor-critic model aims at learning the optimal weights in a continuous action space. An extensive empirical study on various real-world datasets demonstrates that our method achieves excellent or on par results in comparison to the state-of-the-art approaches as well as several baselines.

The use of machine learning methods to inform consequential decisions is increasingly expanding across many fields. As a result, the ability to interpret these models has become to a greater extent crucial to increase the related-technologies acceptance level and reliability. In this paper, we propose an active sampling approach for learning accurately interpretable surrogate machine learning model to better approximate black-box models for supervised learning problems. Hence, the surrogate model is used to learn the black-box model and reflect its properties. Active sampling is used as an informed sampling method to adaptively and iteratively build an optimized training set based on the predictions of the black-box model to enhance the accuracy of the surrogate model. Subsequently, the surrogate model is used to interpret and debug the black-box model. The developed method is flexible and can be used to approximate any family of black-box models using any type of interpretable machine learning models, as it only requires the ability to compute their outputs. It is also applicable to both regression and classification tasks. In this work, we bring focus to decision tree due to their proven high interpretability. An experimental evaluation of the method on several real-world data sets is presented to show its flexibility and its robustness compared to traditional approaches for learning surrogate models.

Massive data broadcast by GPS-equipped vehicles provide unprecedented opportunities. One of the main tasks in order to optimize our transportation networks is to build data-driven real-time decision support systems. However, the dynamic environments where the networks operate disallow the traditional assumptions required to put in practice many off-the-shelf supervised learning algorithms, such as finite training sets or stationary distributions. In this paper, we propose BRIGHT: a drift-aware supervised learning framework to predict demand quantities. BRIGHT aims to provide accurate predictions for short-term horizons through a creative ensemble of time series analysis methods that handles distinct types of concept drift. By selecting neighborhoods dynamically, BRIGHT reduces the likelihood of overfitting. By ensuring diversity among the base learners, BRIGHT ensures a high reduction of variance while keeping bias stable. Experiments were conducted using three large-scale heterogeneous real-world transportation networks in Porto (Portugal), Shanghai (China), and Stockholm (Sweden), as well as with controlled experiments using synthetic data where multiple distinct drifts were artificially induced. The obtained results illustrate the advantages of BRIGHT in relation to state-of-the-art methods for this task.

The dynamic behavior of urban mobility patterns makes matching taxi supply with demand as one of the biggest challenges in this industry. Recently, the increasing availability of massive broadcast GPS data has encouraged the exploration of this issue under different perspectives. One possible solution is to build a data-driven real-time taxi-dispatching recommender system. However, existing systems are based on strong assumptions such as stationary demand distributions and finite training sets, which make them inadequate for modeling the dynamic nature of the network. In this paper, we propose BRIGHT: a drift-aware supervised learning framework which aims to provide accurate predictions for short-term horizon taxi demand quantities through a creative ensemble of time series analysis methods that handle distinct types of concept drift. A large experimental set-up which includes three real-world transportation networks and a synthetic test-bed with artificially inserted concept drifts, was employed to illustrate the advantages of BRIGHT when compared to S.o.A methods for this problem.