Search

Working with electronic health records (EHRs) is known to be challenging due to several reasons. These reasons include not having: 1) similar lengths (per visit), 2) the same number of observations (per patient), and 3) complete entries in the available records. These issues hinder the performance of the predictive models created using EHRs. In this paper, we approach these issues by presenting a model for the combined task of imputing and predicting values for the irregularly observed and varying length EHR data with missing entries. Our proposed model (dubbed as Bi-GAN) uses a bidirectional recurrent network in a generative adversarial setting. In this architecture, the generator is a bidirectional recurrent network that receives the EHR data and imputes the existing missing values. The discriminator attempts to discriminate between the actual and the imputed values generated by the generator. Using the input data in its entirety, Bi-GAN learns how to impute missing elements in-between (imputation) or outside of the input time steps (prediction). Our method has three advantages to the state-of-the-art methods in the field: (a) one single model performs both the imputation and prediction tasks; (b) the model can perform predictions using time-series of varying length with missing data; (c) it does not require to know the observation and prediction time window during training and can be used for the predictions with different observation and prediction window lengths, for short- and long-term predictions. We evaluate our model on two large EHR datasets to impute and predict body mass index (BMI) values and show its superior performance in both settings.

Social networks as well as neighborhood environments have been shown to effect obesity-related behaviors including energy intake and physical activity. Accordingly, harnessing social networks to improve targeting of obesity interventions may be promising to the extent this leads to social multiplier effects and wider diffusion of intervention impact on populations. However, the literature evaluating network-based interventions has been inconsistent. Computational methods like agent-based models (ABM) provide researchers with tools to experiment in a simulated environment. We develop an ABM to compare conventional targeting methods (random selection, based on individual obesity risk, and vulnerable areas) with network-based targeting methods. We adapt a previously published and validated model of network diffusion of obesity-related behavior. We then build social networks among agents using a more realistic approach. We calibrate our model first against national-level data. Our results show that network-based targeting may lead to greater population impact. We also present a new targeting method that outperforms other methods in terms of intervention effectiveness at the population level., Highlights • While behavioral interventions exist for obesity, uncertainty remains about how best to target those interventions to maximize population-level impact. • Agent-based models are simulation tools that can be used to evaluate different targeting strategies to help policy makers. • Targeting intervention resources using social networks may lead to greater population impact of behavioral interventions compared to traditional methods of targeting.

With the increasing availability of wearable devices, continuous monitoring of individuals' physiological and behavioral patterns has become significantly more accessible. Access to these continuous patterns about individuals' statuses offers an unprecedented opportunity for studying complex diseases and health conditions such as type 2 diabetes (T2D). T2D is a widely common chronic disease that its roots and progression patterns are not fully understood. Predicting the progression of T2D can inform timely and more effective interventions to prevent or manage the disease. In this study, we have used a dataset related to 63 patients with T2D that includes the data from two different types of wearable devices worn by the patients: continuous glucose monitoring (CGM) devices and activity trackers (ActiGraphs). Using this dataset, we created a model for predicting the levels of four major biomarkers related to T2D after a one-year period. We developed a wide and deep neural network and used the data from the demographic information, lab tests, and wearable sensors to create the model. The deep part of our method was developed based on the long short-term memory (LSTM) structure to process the time-series dataset collected by the wearables. In predicting the patterns of the four biomarkers, we have obtained a root mean square error of ±1.67% for HBA1c, ±6.22 mg/dl for HDL cholesterol, ±10.46 mg/dl for LDL cholesterol, and ±18.38 mg/dl for Triglyceride. Compared to existing models for studying T2D, our model offers a more comprehensive tool for combining a large variety of factors that contribute to the disease.

Urban simulations are an important tool for analyzing many policy questions relating to the usage of public space, roads, and communal transportation; they can be used to predict the long-term impact of new construction projects, traffic restrictions, and zoning laws. However, it is unwise to rely upon predictions from a single model since each technique possesses different strengths and weaknesses and can be highly sensitive to the choice of parameters and initial conditions. In this article, we describe a hybrid approach for combining agent-based and stochastic simulations (Markov chain Monte Carlo, MCMC) to improve the accuracy and reduce the variance of long-term predictions. In our proposed approach, the agent-based model is used to bootstrap the proposal distribution for the MCMC estimator. To demonstrate the applicability of our modeling technique, this article presents a case study describing the usage of our hybrid simulation method for forecasting transportation patterns and parking lot utilization on a large university campus. A comparison of our simulation results against an independently collected dataset reveals that our hybrid approach accurately predicts parking lot usage and performs significantly better than other comparable modeling techniques. Developing novel architectures for combining the predictions of agent-based models can produce insights that are different than simply selecting the best model.

Holonic multi-agent systems are a special category of multi-agent systems that best fit to environments with numerous agents and high complexity. Like in general multi-agent systems, the agents in the holonic system may negotiate with each other. These systems have their own characteristics and structure, for which a specific negotiation mechanism is required. This mechanism should be simple, fast and operable in real world applications. It would be better to equip negotiators with a learning method which can efficiently use the available information. The learning method should itself be fast, too. Additionally, this mechanism should match the special characteristics of the holonic multi-agent systems. In this paper, we introduce such a negotiation method. Experimental results demonstrate the efficiency of this new approach.

Although there are many explicit collaboration mechanisms in which human teams commit to a joint project, a high level of implicit coordination can be achieved in human societies through the propagation of social norms. A norm can be defined as a “a behavioral rule that is considered valid by the majority of the population” [1]. Norms play a significant role in determining the behavior of people in human societies, and have been successfully used to achieve coordination within normative multi-agent systems.

Agent-based models are a useful technique for rapidly prototyping complex social systems, they are widely used in a number of disciplines and can yield theoretical insights that are different from those produced by a variable based analysis. However, it remains difficult to compare the results of two models and to validate the performance of an agent-based simulation. In this paper, we present a case study on how to analyze the relationship between agent-based models using category theory. Category theory is a powerful mathematical methodology that was originally introduced to organize mathematical ideas according to their shared structure. It has been successfully employed in abstract mathematical domains, but has also enjoyed some success as a formalism for software engineering. Here we present a procedure for analyzing agent-based models using category theory and a case study in its usage at analyzing two different types of simulations.

The Markov Chain Monte Carlo (MCMC) family of methods form a valuable part of the toolbox of social modeling and prediction techniques, enabling modelers to generate samples and summary statistics of a population of interest with minimal information. It has been used successfully to model changes over time in many types of social systems, including patterns of disease spread, adolescent smoking, and geopolitical conflicts. In MCMC an initial proposal distribution is iteratively refined until it approximates the posterior distribution. However, the selection of the proposal distribution can have a significant impact on model convergence. In this paper, we propose a new hybrid modeling technique in which an agent-based model is used to initialize the proposal distribution of the MCMC simulation. We demonstrate the use of our modeling technique in an urban transportation prediction scenario and show that the hybrid combined model produces more accurate predictions than either of the parent models.

Due to their cheap development costs and ease of deployment, surveys and questionnaires are useful tools for gathering information about the activity patterns of a large group and can serve as a valuable supplement to tracking studies done with mobile devices. However in raw form, general survey data is not necessarily useful for answering predictive questions about the behavior of a large social system. In this paper, we describe a method for generating agent activity profiles from survey data for an agent-based model (ABM) of transportation patterns of 47,000 students on a university campus. We compare the performance of our agent-based model against a Markov Chain Monte Carlo (MCMC) simulation based directly on the distributions fitted from the survey data. A comparison of our simulation results against an independently collected dataset reveals that our ABM can be used to accurately forecast parking behavior over the semester and is significantly more accurate than the MCMC estimator.

Any rational agent involving in a multi-agent systems negotiation tries to optimize the negotiation outcome based on its interests or utility function. Negotiations in multi-agent systems are usually complex, and a lot of variables exist which affect the agents' decisions. This becomes more visible in competitive or multi-issue types of negotiations. So, the negotiator agents need an efficient mechanism to do well. The key solution to this type of problems is employing a powerful and operative learning method. An agent tries to learn information it obtains from its environment in order to make the best decisions during the negotiations. In real-world multi-agent negotiations, the main source of usable data is the negotiators' behaviors. So, a good learning approach should be able to extract the buried information in the 'negotiation history'. In this work, we used an ARTMAP artificial neural network as a powerful and efficient learning tool. The main role of this component is to predict other agents' actions/offers in the next rounds of negotiation. When an agent finds out what are the most possible offers which will be proposed, it can predict the outcomes of its decisions. In addition, a new method to apply this information and determine next moves in a negotiation is proposed. The obtained experimental results show that this method can be used effectively in real multi-agent negotiations.

Negotiations are among the most common ways that agents in a multi-agent system use to reach agreements. Because negotiations commonly are multi-lateral and multi-issue, these processes become more difficult. In the real world applications this becomes more important where the autonomous agents involved in a negotiation should reach maximum payoff in minimum time. In this work a new negotiation mechanism is proposed that is based on the multi-objective genetic algorithms. Several measures are defined that can show fitness of an offer in the set of feasible offers that an agent can have in each round of negotiations. The results show that this method can be used in real applications and is competitive with existing approaches.