Personalized Active Learning for Activity Classification in Wireless Health

Abstract— Enabling accurate and low-cost classification of a range of motion activities is important for a variety of applications, ranging from treatment to rehabilitation to training. This paper proposes a novel contextual online learning method for activity classification based on data captured by lowcost, body-worn intertial sensors and smartphones. The proposed method is able to address the unique challenges arising in online monitoring and activity classification with requirements for personalization and online adaptation needing to be fulfilled without a training phase. Another key challenge of activity classification is that the labels may change over time, as the data as well as the activity to be monitored evolve continuously, and the true label is often costly and difficult to obtain. The proposed algorithm is able to actively learn when to ask for the true label by assessing the benefits and costs of obtaining them. We rigorously characterize the performance of the proposed learning algorithm and prove that the learning regret (i.e. performance loss due to learning as compared to an omniscient oracle) is sublinear in time, thereby ensuring fast convergence to the optimal reward as well as providing shortterm performance guarantees. Our experiments show that the proposed algorithm outperforms existing algorithms (e.g. online weighted majority and online AdaBoost) in terms of both providing higher classification accuracy as well as lower energy consumption. Enabling accurate and low-cost classification of a range of motion activities is important for a variety of applications, ranging from treatment to rehabilitation to training. This paper proposes a novel contextual online learning method for activity classification based on data captured by lowcost, body-worn intertial sensors and smartphones. The proposed method is able to address the unique challenges arising in online monitoring and activity classification with requirements for personalization and online adaptation needing to be fulfilled without a training phase. Another key challenge of activity classification is that the labels may change over time, as the data as well as the activity to be monitored evolve continuously, and the true label is often costly and difficult to obtain. The proposed algorithm is able to actively learn when to ask for the true label by assessing the benefits and costs of obtaining them. We rigorously characterize the performance of the proposed learning algorithm and prove that the learning regret (i.e. performance loss due to learning as compared to an omniscient oracle) is sublinear in time, thereby ensuring fast convergence to the optimal reward as well as providing shortterm performance guarantees. Our experiments show that the proposed algorithm outperforms existing algorithms (e.g. online weighted majority and online AdaBoost) in terms of both providing higher classification accuracy as well as lower energy consumption.