Fast reinforcement learning with generalized policy updates.

The combination of reinforcement learning with deep learning is a promising approach to tackle important sequential decision-making problems that are currently intractable. One obstacle to overcome is the amount of data needed by learning systems of this type. In this article, we propose to address this issue through a divide-and-conquer approach. We argue that complex decision problems can be naturally decomposed into multiple tasks that unfold in sequence or in parallel. By associating each task with a reward function, this problem decomposition can be seamlessly accommodated within the standard reinforcement-learning formalism. The specific way we do so is through a generalization of two fundamental operations in reinforcement learning: policy improvement and policy evaluation. The generalized version of these operations allow one to leverage the solution of some tasks to speed up the solution of others. If the reward function of a task can be well approximated as a linear combination of the reward functions of tasks previously solved, we can reduce a reinforcement-learning problem to a simpler linear regression. When this is not the case, the agent can still exploit the task solutions by using them to interact with and learn about the environment. Both strategies considerably reduce the amount of data needed to solve a reinforcement-learning problem.

Automatic identification of Web-based risk markers for health events.

BACKGROUND: The escalating cost of global health care is driving the development of new technologies to identify early indicators of an individual's risk of disease. Traditionally, epidemiologists have identified such risk factors using medical databases and lengthy clinical studies but these are often limited in size and cost and can fail to take full account of diseases where there are social stigmas or to identify transient acute risk factors. OBJECTIVE: Here we report that Web search engine queries coupled with information on Wikipedia access patterns can be used to infer health events associated with an individual user and automatically generate Web-based risk markers for some of the common medical conditions worldwide, from cardiovascular disease to sexually transmitted infections and mental health conditions, as well as pregnancy. METHODS: Using anonymized datasets, we present methods to first distinguish individuals likely to have experienced specific health events, and classify them into distinct categories. We then use the self-controlled case series method to find the incidence of health events in risk periods directly following a user's search for a query category, and compare to the incidence during other periods for the same individuals. RESULTS: Searches for pet stores were risk markers for allergy. We also identified some possible new risk markers; for example: searching for fast food and theme restaurants was associated with a transient increase in risk of myocardial infarction, suggesting this exposure goes beyond a long-term risk factor but may also act as an acute trigger of myocardial infarction. Dating and adult content websites were risk markers for sexually transmitted infections, such as human immunodeficiency virus (HIV). CONCLUSIONS: Web-based methods provide a powerful, low-cost approach to automatically identify risk factors, and support more timely and personalized public health efforts to bring human and economic benefits.

Detecting disease outbreaks in mass gatherings using Internet data.

BACKGROUND: Mass gatherings, such as music festivals and religious events, pose a health care challenge because of the risk of transmission of communicable diseases. This is exacerbated by the fact that participants disperse soon after the gathering, potentially spreading disease within their communities. The dispersion of participants also poses a challenge for traditional surveillance methods. The ubiquitous use of the Internet may enable the detection of disease outbreaks through analysis of data generated by users during events and shortly thereafter. OBJECTIVE: The intent of the study was to develop algorithms that can alert to possible outbreaks of communicable diseases from Internet data, specifically Twitter and search engine queries. METHODS: We extracted all Twitter postings and queries made to the Bing search engine by users who repeatedly mentioned one of nine major music festivals held in the United Kingdom and one religious event (the Hajj in Mecca) during 2012, for a period of 30 days and after each festival. We analyzed these data using three methods, two of which compared words associated with disease symptoms before and after the time of the festival, and one that compared the frequency of these words with those of other users in the United Kingdom in the days following the festivals. RESULTS: The data comprised, on average, 7.5 million tweets made by 12,163 users, and 32,143 queries made by 1756 users from each festival. Our methods indicated the statistically significant appearance of a disease symptom in two of the nine festivals. For example, cough was detected at higher than expected levels following the Wakestock festival. Statistically significant agreement (chi-square test, P<.01) between methods and across data sources was found where a statistically significant symptom was detected. Anecdotal evidence suggests that symptoms detected are indeed indicative of a disease that some users attributed to being at the festival. CONCLUSIONS: Our work shows the feasibility of creating a public health surveillance system for mass gatherings based on Internet data. The use of multiple data sources and analysis methods was found to be advantageous for rejecting false positives. Further studies are required in order to validate our findings with data from public health authorities.