Cherian A.,University of Minnesota |
Sra S.,Empirical |
Banerjee A.,University of Minnesota |
Papanikolopoulos N.,University of Minnesota
IEEE Transactions on Pattern Analysis and Machine Intelligence | Year: 2013
Covariance matrices have found success in several computer vision applications, including activity recognition, visual surveillance, and diffusion tensor imaging. This is because they provide an easy platform for fusing multiple features compactly. An important task in all of these applications is to compare two covariance matrices using a (dis)similarity function, for which the common choice is the Riemannian metric on the manifold inhabited by these matrices. As this Riemannian manifold is not flat, the dissimilarities should take into account the curvature of the manifold. As a result, such distance computations tend to slow down, especially when the matrix dimensions are large or gradients are required. Further, suitability of the metric to enable efficient nearest neighbor retrieval is an important requirement in the contemporary times of big data analytics. To alleviate these difficulties, this paper proposes a novel dissimilarity measure for covariances, the Jensen-Bregman LogDet Divergence (JBLD). This divergence enjoys several desirable theoretical properties and at the same time is computationally less demanding (compared to standard measures). Utilizing the fact that the square root of JBLD is a metric, we address the problem of efficient nearest neighbor retrieval on large covariance datasets via a metric tree data structure. To this end, we propose a K-Means clustering algorithm on JBLD. We demonstrate the superior performance of JBLD on covariance datasets from several computer vision applications. © 1979-2012 IEEE.
Burke C.J.,University of Cambridge |
Burke C.J.,Empirical |
Tobler P.N.,University of Cambridge |
Baddeley M.,University of Cambridge |
Schultz W.,University of Cambridge
Proceedings of the National Academy of Sciences of the United States of America | Year: 2010
Individuals can learn by interacting with the environment and experiencing a difference between predicted and obtained outcomes (prediction error). However, many species also learn by observing the actions and outcomes of others. In contrast to individual learning, observational learning cannot be based on directly experienced outcome prediction errors. Accordingly, the behavioral and neural mechanisms of learning through observation remain elusive. Here we propose that human observational learning can be explained by two previously uncharacterized forms of prediction error, observational action prediction errors (the actual minus the predicted choice of others) and observational outcome prediction errors (the actual minus predicted outcome received by others). In a functional MRI experiment, we found that brain activity in the dorsolateral prefrontal cortex and the ventromedial prefrontal cortex respectively corresponded to these two distinct observational learning signals.
Popp D.,Syracuse University |
Popp D.,National Bureau of Economic Research |
Hascic I.,Empirical |
Medhi N.,Syracuse University
Energy Economics | Year: 2011
We consider investment in wind, solar photovoltaic, geothermal, and electricity from biomass and waste across 26 OECD countries from 1991 to 2004. Using the PATSTAT database, we obtain a comprehensive list of patents for each of these technologies throughout the world, which we use to assess the impact of technological change on investment in renewable energy capacity. We consider four alternative methods for counting patents, using two possible filters: weighting patents by patent family size and including only patent applications filed in multiple countries. For each patent count, we create knowledge stocks representing the global technological frontier. We find that technological advances do lead to greater investment, but the effect is small. Investments in other carbon-free energy sources, such as hydropower and nuclear power, serve as substitutes for renewable energy. Comparing the effectiveness of our four patent counts, we find that both using only patents filed in multiple countries and weighting by family size improve the fit of the model. © 2010 Elsevier B.V.
Kober J.,Bielefeld University |
Kober J.,Honda Research Institute Europe |
Bagnell J.A.,Carnegie Mellon University |
Peters J.,Empirical |
Peters J.,TU Darmstadt
International Journal of Robotics Research | Year: 2013
Reinforcement learning offers to robotics a framework and set of tools for the design of sophisticated and hard-to-engineer behaviors. Conversely, the challenges of robotic problems provide both inspiration, impact, and validation for developments in reinforcement learning. The relationship between disciplines has sufficient promise to be likened to that between physics and mathematics. In this article, we attempt to strengthen the links between the two research communities by providing a survey of work in reinforcement learning for behavior generation in robots. We highlight both key challenges in robot reinforcement learning as well as notable successes. We discuss how contributions tamed the complexity of the domain and study the role of algorithms, representations, and prior knowledge in achieving these successes. As a result, a particular focus of our paper lies on the choice between model-based and model-free as well as between value-function-based and policy-search methods. By analyzing a simple problem in some detail we demonstrate how reinforcement learning approaches may be profitably applied, and we note throughout open questions and the tremendous potential for future research. © The Author(s) 2013.
Johnstone N.,Empirical |
Hascic I.,Empirical |
Popp D.,Syracuse University
Environmental and Resource Economics | Year: 2010
This paper examines the effect of environmental policies on technological innovation in the specific case of renewable energy. The analysis is conducted using patent data on a panel of 25 countries over the period 1978-2003. We find that public policy plays a significant role in determining patent applications. Different types of policy instruments are effective for different renewable energy sources. Broad-based policies, such as tradable energy certificates, are more likely to induce innovation on technologies that are close to competitive with fossil fuels. More targeted subsidies, such as feed-in tariffs, are needed to induce innovation on more costly energy technologies, such as solar power. Springer Science+Business Media B.V. 2009.