Prediction With Unpredictable Feature Evolution

Open-Ended Online Learning for Autonomous Visual Perception

The visual perception systems aim to autonomously collect consecutive visual data and perceive the relevant information online like human beings. In comparison with the classical static visual systems focusing on fixed tasks (e.g., face recognition for visual surveillance), the real-world visual systems (e.g., the robot visual system) often need to handle unpredicted tasks and dynamically changed environments, which need to imitate human-like intelligence with open-ended online learning ability. Therefore, we provide a comprehensive analysis of open-ended online learning problems for autonomous visual perception in this survey. Based on "what to online learn" among visual perception scenarios, we classify the open-ended online learning methods into five categories: instance incremental learning to handle data attributes changing, feature evolution learning for incremental and decremental features with the feature dimension changed dynamically, class incremental learning and task incremental learning aiming at online adding new coming classes/tasks, and parallel and distributed learning for large-scale data to reveal the computational and storage advantages. We discuss the characteristic of each method and introduce several representative works as well. Finally, we introduce some representative visual perception applications to show the enhanced performance when using various open-ended online learning models, followed by a discussion of several future directions.

Open-environment machine learning

Conventional machine learning studies generally assume close-environment scenarios where important factors of the learning process hold invariant. With the great success of machine learning, nowadays, more and more practical tasks, particularly those involving open-environment scenarios where important factors are subject to change, called open-environment machine learning in this article, are present to the community. Evidently, it is a grand challenge for machine learning turning from close environment to open environment. It becomes even more challenging since, in various big data tasks, data are usually accumulated with time, like streams, while it is hard to train the machine learning model after collecting all data as in conventional studies. This article briefly introduces some advances in this line of research, focusing on techniques concerning emerging new classes, decremental/incremental features, changing data distributions and varied learning objectives, and discusses some theoretical issues.

IoT and Satellite Sensor Data Integration for Assessment of Environmental Variables: A Case Study on NO2

This paper introduces a novel approach to increase the spatiotemporal resolution of an arbitrary environmental variable. This is achieved by utilizing machine learning algorithms to construct a satellite-like image at any given time moment, based on the measurements from IoT sensors. The target variables are calculated by an ensemble of regression models. The observed area is gridded, and partitioned into Voronoi cells based on the IoT sensors, whose measurements are available at the considered time. The pixels in each cell have a separate regression model, and take into account the measurements of the central and neighboring IoT sensors. The proposed approach was used to assess NO2 data, which were obtained from the Sentinel-5 Precursor satellite and IoT ground sensors. The approach was tested with three different machine learning algorithms: 1-nearest neighbor, linear regression and a feed-forward neural network. The highest accuracy yield was from the prediction models built with the feed-forward neural network, with an RMSE of 15.49 ×10-6 mol/m2.