The use of reinforcement learning (RL) to learn policies for just-in-time adaptive interventions (JITAIs) is of significant interest in many behavioral intervention domains including improving levels of physical activity. In a messaging-based physical activity JITAI, a mobile health app is typically used to send messages to a participant to encourage engagement in physical activity. In this setting, RL methods can be used to learn what intervention options to provide to a participant in different contexts. However, deploying RL methods in real physical activity adaptive interventions comes with challenges: the cost and time constraints of real intervention studies result in limited data to learn adaptive intervention policies. Further, commonly used RL simulation environments have dynamics that are of limited relevance to physical activity adaptive interventions and thus shed little light on what RL methods may be optimal for this challenging application domain. In this paper, we introduce StepCountJITAI, an RL environment designed to foster research on RL methods that address the significant challenges of policy learning for adaptive behavioral interventions.
Physical activity helps people maintain a healthy weight and reduces the risk for several chronic diseases. Although this knowledge is widely recognized, adults and children in many countries around the world do not get recommended amounts of physical activity. While many interventions are found to be ineffective at increasing physical activity or reaching inactive populations, there have been anecdotal reports of increased physical activity due to novel mobile games that embed game play in the physical world. The most recent and salient example of such a game is Pokémon Go, which has reportedly reached tens of millions of users in the US and worldwide. We study the effect of Pokémon Go on physical activity through a combination of signals from large-scale corpora of wearable sensor data and search engine logs for 32 thousand users over a period of three months. Pokémon Go players are identified through search engine queries and activity is measured through accelerometry. We find that Pokémon Go leads to significant increases in physical activity over a period of 30 days, with particularly engaged users (i.e., those making multiple search queries for details about game usage) incre
Mobile health applications show promise for scalable physical activity promotion but are often insufficiently personalized. In contrast, health coaching offers highly personalized support but can be prohibitively expensive and inaccessible. This study draws inspiration from health coaching to explore how large language models (LLMs) might address personalization challenges in mobile health. We conduct formative interviews with 12 health professionals and 10 potential coaching recipients to develop design principles for an LLM-based health coach. We then built GPTCoach, a chatbot that implements the onboarding conversation from an evidence-based coaching program, uses conversational strategies from motivational interviewing, and incorporates wearable data to create personalized physical activity plans. In a lab study with 16 participants using three months of historical data, we find promising evidence that GPTCoach gathers rich qualitative information to offer personalized support, with users feeling comfortable sharing concerns. We conclude with implications for future research on LLM-based physical activity support.
This paper addresses the methodology for the quarterly estimation of Compensation of Employees paid by the General Government (GG) sector, in accordance with the European System of Accounts (ESA 2010). Due to the limited high-frequency data availability and the need to guarantee the consistency with annual constraints, quarterly estimation relies on indirect temporal disaggregation techniques. These methods use specific infra-annual indicators as proxies for the variables being estimated. The specific case of the quarterly estimation of Compensation of employees presents several additional challenges. Firstly, the information provided by the sources, based on cash or legal-accrual data, is elaborated to define indicators which respect the accrual ESA 2010 principle as the annual estimates, based on more compliant data sources such as final budgets of public entities. Secondly, at a quarterly level the extraordinary events - such as the recording of delayed collective bargaining agreements which result in arrears - have a strong impact on quarterly indicators, whereas their effect is mitigated at annual level. To attribute these flows to the period when the work is performed, multi-
In this work, we have developed a framework for synthesizing data driven controllers for a class of uncertain switched systems arising in an application to physical activity interventions. In particular, we present an application of probabilistic model predictive control to design an efficient, tractable, and adaptive intervention using behavioral data sets i.e. physical activity behavior. The models of physical activity for each individual are provided for the design of controllers that maximize the probability of achieving a desired physical activity goal subject to intervention specifications. We have tailored the mixed-integer programming-based approach for evaluating the Model Predictive Control decision at each time step.
Activity cliff prediction - identifying positions where small structural changes cause large potency shifts - has been a persistent challenge in computational medicinal chemistry. This work focuses on a parsimonious definition: which small modifications, at which positions, confer the highest probability of an outcome change. Position-level sensitivity is calculated using 25 million matched molecular pairs from 50 ChEMBL targets across six protein families, revealing that two questions have fundamentally different answers. "Which positions vary most?" is answered by scaffold size alone (NDCG@3 = 0.966), requiring no machine learning. "Which are true activity cliffs?" - where small modifications cause disproportionately large effects, as captured by SALI normalization - requires an 11-feature model with 3D pharmacophore context (NDCG@3 = 0.910 vs. 0.839 random), generalizing across all six protein families, novel scaffolds (0.913), and temporal splits (0.878). The model identifies the cliff-prone position first 53% of the time (vs. 27% random - 2x lift), reducing positions a chemist must explore from 3.1 to 2.1 - a 31% reduction in first-round experiments. Predicting which modificat
Human behaviour is heterogeneous and temporally fluctuates. Many studies have focused on inter-event time (IET) fluctuations and have reported that the IET distributions have a long-tailed distribution, which cannot be explained by a stationary Poisson point process. Such phenomenon observed in IET distributions are known as burstiness. Burstiness has also been reported for human physical activity, but the mechanism underlying it has not been clarified. In this study, we collected human physical activity data while specifying the age of the subjects and their situations (for example, children's play and adults' housework), and we analysed their event time-series data. We confirmed the burstiness in both children and adults. For the first time, burstiness studied in physical activities of children between the ages 2 and 5. We also confirmed that the characteristics of the IET distribution are unique to each activity situation. Our results may be critical in the identification of the burstiness mechanisms in human physical activity.
In the realm of Human Activity Recognition (HAR), obtaining high quality and variance data is still a persistent challenge due to high costs and the inherent variability of real-world activities. This study introduces a generation dataset by deep learning approaches (Attention Autoencoder and conditional Generative Adversarial Networks). Another problem that data heterogeneity is a critical challenge, one of the solutions is to shuffle the data to homogenize the distribution. Experimental results demonstrate that the random sequence strategy significantly improves classification performance, achieving an accuracy of up to 0.70 $\pm$ 0.03 and a macro F1 score of 0.64 $\pm$ 0.01. For that, disrupting temporal dependencies through random sequence reordering compels the model to focus on instantaneous recognition, thereby improving robustness against activity transitions. This approach not only broadens the effective training dataset but also offers promising avenues for enhancing HAR systems in complex, real-world scenarios.
The continuous wavelet transform is adapted to account for signal truncation through renormalization and by modifying the shape of the analyzing window. Comparison is made of the instant and integrated wavelet power with previous algorithms. The edge adapted and renormalized admissible wavelet transforms are used to estimate the level of solar magnetic activity from the sunspot record. The solar activity is compared to Oerlemans' temperature reconstruction and to the Central England Temperature record. A correlation is seen for years between 1610 and 1990, followed by a strong deviation as the recently observed temperature increases.
The proliferation of deep learning has significantly advanced various fields, yet Human Activity Recognition (HAR) has not fully capitalized on these developments, primarily due to the scarcity of labeled datasets. Despite the integration of advanced Inertial Measurement Units (IMUs) in ubiquitous wearable devices like smartwatches and fitness trackers, which offer self-labeled activity data from users, the volume of labeled data remains insufficient compared to domains where deep learning has achieved remarkable success. Addressing this gap, in this paper, we propose a novel approach to improve wearable sensor-based HAR by introducing a pose-to-sensor network model that generates sensor data directly from 3D skeleton pose sequences. our method simultaneously trains the pose-to-sensor network and a human activity classifier, optimizing both data reconstruction and activity recognition. Our contributions include the integration of simultaneous training, direct pose-to-sensor generation, and a comprehensive evaluation on the MM-Fit dataset. Experimental results demonstrate the superiority of our framework with significant performance improvements over baseline methods.
Complex physical simulations often require trade-offs between model fidelity and computational feasibility. We introduce Adaptive Online Emulation (AOE), which dynamically learns neural network surrogates during simulation execution to accelerate expensive components. Unlike existing methods requiring extensive offline training, AOE uses Online Sequential Extreme Learning Machines (OS-ELMs) to continuously adapt emulators along the actual simulation trajectory. We employ a numerically stable variant of the OS-ELM using cumulative sufficient statistics to avoid matrix inversion instabilities. AOE integrates with time-stepping frameworks through a three-phase strategy balancing data collection, updates, and surrogate usage, while requiring orders of magnitude less training data than conventional surrogate approaches. Demonstrated on a 1D atmospheric model of exoplanet GJ1214b, AOE achieves 11.1 times speedup (91% time reduction) across 200,000 timesteps while maintaining accuracy, potentially making previously intractable high-fidelity time-stepping simulations computationally feasible.
Background: Software development teams are increasingly diverse, embedded, and cross-disciplinary. Domain experts (DEs) from different disciplines collaborate with professional software developers (SDEs), bringing complementary expertise in creating and maintaining complex production software. However, contested expectations, divergent problem-solving perspectives, and conflicting priorities lead to friction. Aims: This study aims to investigate the dynamics of emerging collaboration of cross-disciplinary software development (CDSD) by exploring the expectations held by DEs and SDEs and understanding how these frictions manifest in practice. Method: We utilize Activity Theory (AT), a well-established socio-technical framework, as an analytical lens in a grounded, empirical investigation, conducted through a mixed-method study involving 24 interviews (12 DEs and 12 SDEs) and a large-scale validation survey with 293 participants (161 DEs and 132 SDEs). Results: We conceptualize and empirically ground the CDSD dynamics. We identified eight expectations held by SDEs and six by DEs. By mapping these expectations to AT components, we revealed 21 frictions in CDSD and illustrated where an
In the human activity recognition research area, prior studies predominantly concentrate on leveraging advanced algorithms on public datasets to enhance recognition performance, little attention has been paid to executing real-time kitchen activity recognition on energy-efficient, cost-effective edge devices. Besides, the prevalent approach of segregating data collection and context extraction across different devices escalates power usage, latency, and user privacy risks, impeding widespread adoption. This work presents a multi-modal wearable edge computing system for human activity recognition in real-time. Integrating six different sensors, ranging from inertial measurement units (IMUs) to thermal cameras, and two different microcontrollers, this system achieves end-to-end activity recognition, from data capture to context extraction, locally. Evaluation in an unmodified realistic kitchen validates its efficacy in recognizing fifteen activities, including a null class. Employing a compact machine learning model (184.5 kbytes) yields an average accuracy of 87.83 \%, with model inference completed in 25.26 ms on the microcontroller. Comparative analysis with alternative microcontr
This paper presents two models to address the problem of multi-person activity recognition using ambient sensors in a home. The first model, Seq2Res, uses a sequence generation approach to separate sensor events from different residents. The second model, BiGRU+Q2L, uses a Query2Label multi-label classifier to predict multiple activities simultaneously. Performances of these models are compared to a state-of-the-art model in different experimental scenarios, using a state-of-the-art dataset of two residents in a home instrumented with ambient sensors. These results lead to a discussion on the advantages and drawbacks of resident separation and multi-label classification for multi-person activity recognition.
Wrist accelerometers for assessing hallmark measures of physical activity (PA) are rapidly growing with the advent of smartwatch technology. Given the growing popularity of wrist-worn accelerometers, there needs to be a rigorous evaluation for recognizing (PA) type and estimating energy expenditure (EE) across the lifespan. Participants (66% women, aged 20-89 yrs) performed a battery of 33 daily activities in a standardized laboratory setting while a tri-axial accelerometer collected data from the right wrist. A portable metabolic unit was worn to measure metabolic intensity. We built deep learning networks to extract spatial and temporal representations from the time-series data, and used them to recognize PA type and estimate EE. The deep learning models resulted in high performance; the F1 score was: 0.82, 0.81, and 95 for recognizing sedentary, locomotor, and lifestyle activities, respectively. The root mean square error was 1.1 (+/-0.13) for the estimation of EE.
Developing human activity recognition (HAR) systems for smart homes is not straightforward due to varied layouts of the homes and their personalized settings, as well as idiosyncratic behaviors of residents. As such, off-the-shelf HAR systems are effective in limited capacity for an individual home, and HAR systems often need to be derived "from scratch", which comes with substantial efforts and often is burdensome to the resident. Previous work has successfully targeted the initial phase. At the end of this initial phase, we identify seed points. We build on bootstrapped HAR systems and introduce an effective updating and extension procedure for continuous improvement of HAR systems with the aim of keeping up with ever changing life circumstances. Our method makes use of the seed points identified at the end of the initial bootstrapping phase. A contrastive learning framework is trained using these seed points and labels obtained for the same. This model is then used to improve the segmentation accuracy of the identified prominent activities. Improvements in the activity recognition system through this procedure help model the majority of the routine activities in the smart home.
This chapter aims to aid the development of Cyber-Physical Systems (CPS) in automated understanding of events and activities in various applications of video-surveillance. These events are mostly captured by drones, CCTVs or novice and unskilled individuals on low-end devices. Being unconstrained, these videos are immensely challenging due to a number of quality factors. We present an extensive account of the various approaches taken to solve the problem over the years. This ranges from methods as early as Structure from Motion (SFM) based approaches to recent solution frameworks involving deep neural networks. We show that the long-term motion patterns alone play a pivotal role in the task of recognizing an event. Consequently each video is significantly represented by a fixed number of key-frames using a graph-based approach. Only the temporal features are exploited using a hybrid Convolutional Neural Network (CNN) + Recurrent Neural Network (RNN) architecture. The results we obtain are encouraging as they outperform standard temporal CNNs and are at par with those using spatial information along with motion cues. Further exploring multistream models, we conceive a multi-tier fus
Entanglement detection is an important problem in quantum information theory because quantum entanglement is a key resource in quantum information processing. Realignment criteria is a powerful tool for detection of entangled states in bipartite and multipartite quantum system. It is an important criteria for entanglement detection because it works well; not only for negative partial transpose entangled states (NPTES) but also for positive partial transpose entangled states (PPTES). Since the matrix corresponding to realignment map is indefinite so the experimental implementation of the map is an obscure task. In this work, firstly, we have approximated the realignment map to a positive map using the method of structural physical approximation (SPA) and then we have shown that the structural physical approximation of realignment map (SPA-R) is completely positive. Positivity of the constructed map is characterized using moments which can be physically measured. Next, we develop a separability criterion based on our SPA-R map in the form of an inequality and have shown that the developed criterion not only detect NPTES but also PPTES. We have provided some examples to support the re
Finding solar-analog stars with fundamental properties as close as possible to the Sun and studying the characteristics of their surface magnetic activity is a very promising way to understand the solar variability and its associated dynamo process. However, the identification of solar-analog stars depends on the accuracy of the estimated stellar parameters. Thanks to the photometric CoROT and Kepler space missions, the addition of asteroseismic data was proven to provide the most accurate fundamental properties that can be derived from stellar modeling today. Here, we present our latest results on the solar-stellar connection by studying 18 solar analogs that we identified among the Kepler seismic sample (Salabert et al., 2016a). We measured their magnetic activity properties using the observations collected by the Kepler satellite and the ground-based, high-resolution HERMES spectrograph. The photospheric (Sph) and chromospheric (S) magnetic activity proxies of these seismic solar analogs are compared in relation to the solar activity. We show that the activity of the Sun is comparable to the activity of the seismic solar analogs, within the maximum-to-minimum temporal variations
Presented here is a review of present knowledge of the long-term behavior of solar activity on a multi-millennial timescale, as reconstructed using the indirect proxy method. The concept of solar activity is discussed along with an overview of the special indices used to quantify different aspects of variable solar activity, with special emphasis upon sunspot number. Over long timescales, quantitative information about past solar activity can only be obtained using a method based upon indirect proxies, such as the cosmogenic isotopes \super{14}C and \super{10}Be in natural stratified archives (e.g., tree rings or ice cores). We give an historical overview of the development of the proxy-based method for past solar-activity reconstruction over millennia, as well as a description of the modern state. Special attention is paid to the verification and cross-calibration of reconstructions. It is argued that this method of cosmogenic isotopes makes a solid basis for studies of solar variability in the past on a long timescale (centuries to millennia) during the Holocene. A separate section is devoted to reconstructions of strong solar energetic-particle (SEP) events in the past, that sug