The Experts below are selected from a list of 324 Experts worldwide ranked by ideXlab platform

Kristen Grauman - One of the best experts on this subject based on the ideXlab platform.

  • sidekick Policy learning for active visual exploration
    European Conference on Computer Vision, 2018
    Co-Authors: Santhosh K. Ramakrishnan, Kristen Grauman
    Abstract:

    We consider an active visual exploration scenario, where an agent must intelligently select its camera motions to efficiently reconstruct the full environment from only a limited set of narrow field-of-view glimpses. While the agent has full observability of the environment during training, it has only partial observability once deployed, being constrained by what portions it has seen and what camera motions are permissible. We introduce sidekick Policy learning to capitalize on this imbalance of observability. The main idea is a preparatory learning phase that attempts simplified versions of the eventual exploration task, then guides the agent via reward shaping or Initial Policy supervision. To support interpretation of the resulting policies, we also develop a novel Policy visualization technique. Results on active visual exploration tasks with \(360^{\circ }\) scenes and 3D objects show that sidekicks consistently improve performance and convergence rates over existing methods. Code, data and demos are available (Project website: http://vision.cs.utexas.edu/projects/sidekicks/).

  • sidekick Policy learning for active visual exploration
    arXiv: Computer Vision and Pattern Recognition, 2018
    Co-Authors: Santhosh K. Ramakrishnan, Kristen Grauman
    Abstract:

    We consider an active visual exploration scenario, where an agent must intelligently select its camera motions to efficiently reconstruct the full environment from only a limited set of narrow field-of-view glimpses. While the agent has full observability of the environment during training, it has only partial observability once deployed, being constrained by what portions it has seen and what camera motions are permissible. We introduce sidekick Policy learning to capitalize on this imbalance of observability. The main idea is a preparatory learning phase that attempts simplified versions of the eventual exploration task, then guides the agent via reward shaping or Initial Policy supervision. To support interpretation of the resulting policies, we also develop a novel Policy visualization technique. Results on active visual exploration tasks with 360 scenes and 3D objects show that sidekicks consistently improve performance and convergence rates over existing methods. Code, data and demos are available.

Santhosh K. Ramakrishnan - One of the best experts on this subject based on the ideXlab platform.

  • sidekick Policy learning for active visual exploration
    European Conference on Computer Vision, 2018
    Co-Authors: Santhosh K. Ramakrishnan, Kristen Grauman
    Abstract:

    We consider an active visual exploration scenario, where an agent must intelligently select its camera motions to efficiently reconstruct the full environment from only a limited set of narrow field-of-view glimpses. While the agent has full observability of the environment during training, it has only partial observability once deployed, being constrained by what portions it has seen and what camera motions are permissible. We introduce sidekick Policy learning to capitalize on this imbalance of observability. The main idea is a preparatory learning phase that attempts simplified versions of the eventual exploration task, then guides the agent via reward shaping or Initial Policy supervision. To support interpretation of the resulting policies, we also develop a novel Policy visualization technique. Results on active visual exploration tasks with \(360^{\circ }\) scenes and 3D objects show that sidekicks consistently improve performance and convergence rates over existing methods. Code, data and demos are available (Project website: http://vision.cs.utexas.edu/projects/sidekicks/).

  • sidekick Policy learning for active visual exploration
    arXiv: Computer Vision and Pattern Recognition, 2018
    Co-Authors: Santhosh K. Ramakrishnan, Kristen Grauman
    Abstract:

    We consider an active visual exploration scenario, where an agent must intelligently select its camera motions to efficiently reconstruct the full environment from only a limited set of narrow field-of-view glimpses. While the agent has full observability of the environment during training, it has only partial observability once deployed, being constrained by what portions it has seen and what camera motions are permissible. We introduce sidekick Policy learning to capitalize on this imbalance of observability. The main idea is a preparatory learning phase that attempts simplified versions of the eventual exploration task, then guides the agent via reward shaping or Initial Policy supervision. To support interpretation of the resulting policies, we also develop a novel Policy visualization technique. Results on active visual exploration tasks with 360 scenes and 3D objects show that sidekicks consistently improve performance and convergence rates over existing methods. Code, data and demos are available.

Ali Heydari - One of the best experts on this subject based on the ideXlab platform.

  • Stability Analysis of Optimal Adaptive Control Under Value Iteration Using a Stabilizing Initial Policy
    IEEE transactions on neural networks and learning systems, 2017
    Co-Authors: Ali Heydari
    Abstract:

    Adaptive optimal control using value iteration initiated from a stabilizing control Policy is theoretically analyzed. The analysis is in terms of stability of the system during the learning stage and includes the system controlled by any fixed control Policy and also by an evolving Policy. A feature of the presented results is finding subsets of the region of attraction. This is done so that if the Initial condition belongs to this region, the entire state trajectory remains within the training region. Therefore, the function approximation results remain reliable, as no extrapolation will be conducted.

Ram A. Cnaan - One of the best experts on this subject based on the ideXlab platform.

  • DOES SOCIAL WORK EDUCATION HAVE AN IMPACT ON SOCIAL Policy PREFERENCES? A THREE-COHORT STUDY
    Journal of Social Work Education, 2005
    Co-Authors: Idit Weiss, John Gal, Ram A. Cnaan
    Abstract:

    This article examines the impact of social work education on the social Policy preferences of social work students through a panel study of 3 cohorts of students at universities in 2 countries—the United States and Israel. The findings of the study indicate that though the Initial Policy preferences of the students at the beginning of their studies at the 3 universities differed, by the end of their studies the students' preferences were similar and supportive of the welfare state model.

Verstraten Paul - One of the best experts on this subject based on the ideXlab platform.

  • IZA COVID-19 crisis response monitoring: short-run labor market impacts of COVID-19, Initial Policy measures and beyond
    The Institute for the Study of Labor (IZA), 2020
    Co-Authors: Ferreira Priscila, Eichhorst Werner, Rinne Ulf, Marx Paul, Böheim René, Leoni Thomas, Cahuc Pierre, Colussi Tommaso, Jongen, Egbert L. W., Verstraten Paul
    Abstract:

    The unprecedented COVID-19 pandemic has a severe impact on societies, economies and labor markets. However, not all countries, socio-economic groups and sectors are equally affected. For example, occupational groups working in sectors where value chains have been disrupted and lockdowns have had direct impacts are affected more heavily, while the slowdown of hiring activities mostly affects young labor market entrants. As a result, there has been a steep increase in unemployment rates in many countries, but not everywhere to the same extent. Part of this difference can be related to the different role and extent of short-time work schemes, which is now being used more widely than during the Great Recession. Some countries have created or expanded these schemes, making coverage less exclusive and benefits more generous, at least temporarily. But short-time work is certainly not a panacea to “flatten the unemployment curve”. Furthermore, next to providing liquidity support to firms, unemployment benefits have been made more generous in many countries. Often, activation principles have also been temporarily reduced. Some countries have increased access to income support to some extent also for non-standard workers, such as temporary agency workers or self-employed workers, on an ad hoc basis. A major change in working conditions is the broad move towards telework arrangements and work from home. Nonetheless, it appears too early to assess the relative success of national strategies to cope with the pandemic and to revitalize the labor market as well as the medium-term fiscal viability of different support measures. Future monitoring will also have to trace policies to cope with the imminent structural changes that might result from the crisis or might be accelerated by the crisis