We present AutoDOViz, an interactive user interface for automated decision optimization (AutoDO) using reinforcement learning (RL). Decision optimization (DO) has classically being practiced by dedicated DO researchers [43] where experts need to spend long periods of time fine tuning a solution through trial-and-error. AutoML pipeline search has sought to make it easier for a data scientist to find the best machine learning pipeline by leveraging automation to search and tune the solution. More recently, these advances have been applied to the domain of AutoDO [36], with a similar goal to find the best reinforcement learning pipeline through algorithm selection and parameter tuning. However, Decision Optimization requires significantly more complex problem specification when compared to an ML problem. AutoDOViz seeks to lower the barrier of entry for data scientists in problem specification for reinforcement learning problems, leverage the benefits of AutoDO algorithms for RL pipeline search and finally, create visualizations and policy insights in order to facilitate the typical interactive nature when communicating problem formulation and solution proposals between DO experts and domain experts. In this paper, we report our findings from semi-structured expert interviews with DO practitioners as well as business consultants, leading to design requirements for human-centered automation for DO with RL. We evaluate a system implementation with data scientists and find that they are significantly more open to engage in DO after using our proposed solution. AutoDOViz further increases trust in RL agent models and makes the automated training and evaluation process more comprehensible. As shown for other automation in ML tasks [33, 59], we also conclude automation of RL for DO can benefit from user and vice-versa when the interface promotes human-in-the-loop.

Optimizing a machine learning (ML) pipeline has been an important topic of AI and ML. Despite recent progress, pipeline optimization remains a challenging problem, due to potentially many combinations to consider as well as slow training and validation. We present the BLDS algorithm for optimized algorithm selection (ML operations) in a fixed ML pipeline structure. BLDS performs multi-fidelity optimization for selecting ML algorithms trained with smaller computational overhead, while controlling its pipeline search based on multi-armed bandit and limited discrepancy search. Our experiments on well-known classification benchmarks show that BLDS is superior to competing algorithms. We also combine BLDS with hyperparameter optimization, empirically showing the advantage of BLDS.

AutoML automatically selects, composes and parameterizes machine learning algorithms into a workflow or pipeline of operations that aims at maximizing performance on a given dataset. Although current methods for AutoML achieved impressive results they mostly concentrate on optimizing fixed linear workflows. In this paper, we take a different approach and focus on generating and optimizing pipelines of complex directed acyclic graph shapes. These complex pipeline structure may lead to discovering hidden features and thus boost performance considerably. We explore the power of heuristic search and context-free grammars to search and optimize these kinds of pipelines. Experiments on various benchmark datasets show that our approach is highly competitive and often outperforms existing AutoML systems.

The growing aging population faces a number of challenges, including rising medical cost, inadequate number of medical doctors and healthcare professionals, as well as higher incidence of misdiagnosis. There is an increasing demand for a better healthcare support for the elderly and one promising solution is the development of a context-aware middleware infrastructure for pervasive health/wellness-care. This allows the accurate and timely delivery of health/medical information among the patients, doctors and healthcare workers through a widespread deployment of wireless sensor networks and mobile devices. In this paper, we present our design and implementation of such a context-aware middleware for pervasive homecare (CAMPH). The middleware offers several key-enabling system services that consist of P2P-based context query processing, context reasoning for activity recognition and context-aware service management. It can be used to support the development and deployment of various homecare services for the elderly such as patient monitoring, location-based emergency response, anomalous daily activity detection, pervasive access to medical data and social networking. We have developed a prototype of the middleware and demonstrated the concept of providing a continuing-care to an elderly with the collaborative interactions spanning multiple physical spaces: person, home, office and clinic. The results of the prototype show that our middleware approach achieves good efficiency of context query processing and good accuracy of activity recognition.

Context-aware computing is a key paradigm of ubiquitous computing in which applications automatically adapt their operations to dynamic context data from multiple sources. Managing a number of distributed sources, a middleware that facilitates the development of context-aware applications must provide a uniform view of all these sources to the applications. Local schemas of context data from individual sources need to be matched into a set of global schemas in the middleware, upon which applications can issue context queries to acquire data. In this paper, we study this problem of schema matching for context-aware computing. We propose a multi-criteria algorithm to determine candidate attribute matches between two schemas. The algorithm adaptively adjusts the priorities of different criteria based on previous matching results to improve the efficiency and accuracy of succeeding operations. We further develop an algorithm to categorize a new local schema into one of the global schemas whenever possible via a shared attribute dictionary. Our results based on schemas from real-world websites demonstrate the good matching accuracy achieved by our algorithms.

Studying the brain as a system requires global collaborations and interdisciplinary approaches which necessitate the development of tools to help scientists in the management, sharing, and synthesis of disparate research resources. Recognizing the benefits and importance of global collaboration and sharing, the INCF (International Neuroinformatics Coordinating Facility) started to coordinate the global effort of establishing different Neuroinformatics (NI) Portals among the participating countries. In Japan, this initiative is starting to take shape through the establishment of the different NI platforms under the coordination of the NIJC (NI Japan Center). Each NI platform in Japan such as "visiome" [http://platform. visiome. org], requires their own set of keywords that represent important terms covering their respective field of study. One important role of this predefined keyword list is to help scientists classify the contents of their contributions and group related resources based on these keywords. It is vital that this predefined list should be properly chosen to cover the necessary areas. Currently, the process of identifying these keywords relies on the availability of human experts which does not scale well considering that the different fields are rapidly evolving. This issue prompted us to develop a new algorithm for a tool to automatically filter terms which are most likely considered as keywords by human experts. We discuss its effectiveness and tested its performance using the abstracts of the Vision Research Journal (VR) as a test case.

Evolving gradient-learning artificial neural networks (ANNs) using an evolutionary algorithm (EA) is a popular approach to address the local optima and design problems of ANN. The typical approach is to combine the strength of backpropagation (BP) in weight learning and EA's capability of searching the architecture space. However, the BP's "gradient descent" approach requires a highly computer-intensive operation that relatively restricts the search coverage of EA by compelling it to use a small population size. To address this problem, we utilized mutation-based genetic neural network (MGNN) to replace BP by using the mutation strategy of local adaptation of evolutionary programming (EP) to effect weight learning. The MGNN's mutation enables the network to dynamically evolve its structure and adapt its weights at the same time. Moreover, MGNN's EP-based encoding scheme allows for a flexible and less restricted formulation of the fitness function and makes fitness computation fast and efficient. This makes it feasible to use larger population sizes and allows MGNN to have a relatively wide search coverage of the architecture space. MGNN implements a stopping criterion where overfitness occurrences are monitored through "sliding-windows" to avoid premature learning and overlearning. Statistical analysis of its performance to some well-known classification problems demonstrate its good generalization capability. It also reveals that locally adapting or scheduling the strategy parameters embedded in each individual network may provide a proper balance between the local and global searching capabilities of MGNN.

One important issue in developing dynamic algorithms that changes the structure and weights of ANN (artificial neural networks) is how to achieve a proper balance between network complexity and its generalization capability. Typical hybrid approaches to address this problem incorporates EA strategy using a population of backpropagation networks. Since individuals undergo backpropagation networks. Since individuals undergo backpropagation training, this approach is inefficient and inherits the pitfalls of gradient learning. SEPA (structure evolution and parameter adaptation) addresses these issues using an encoding scheme where network weights and connections are encoded in matrices of real numbers. Network parameters are locally encoded and undergo local adaptation with fitness evaluation consisting mainly of fast feed-forward matrix operations that can be implemented in parallel or distributed environment. Experimental results show that SEPA's strategy produces optimal network structure with fast convergence, high consistency, and good generalization capability.