Search Constraints

Start Over
Filtering by: Creator Congalton, Russell G. Remove constraint Creator: Congalton, Russell G. Advisor Tumer, Kagan Remove constraint Advisor: Tumer, Kagan

Search Results

  • Learning to coordinate in sparse asymmetric multiagent systems

    Creator:
    Dixit, Gaurav
    Abstract:
    Multiagent learning offers a rich framework to address challenging real-world problems such as remote exploration and healthcare coordination, which require autonomous agents to express elaborate interactions. To be effective in such systems, agents must collectively reason about and pursue high-level, long-term, and possibly nebulous objectives while adapting their strategy to...
    Resource Type:
    Dissertation
    Full Text:
    excavators. This scenarios sees a majority of excavators in the team (over 40%) (G) that learn to either

  • Counterfactual thinking applied to multiagent systems

    Creator:
    Zerbel, Nicholas
    Abstract:
    Multiagent approaches are well suited to designing autonomous solutions for systems that feature complex interactions between many individuals such as in autonomous traffic systems and multi-robot exploration systems. However, creating autonomous agents that function effectively in these systems is a challenging task. In these complex environments, agents need informative reward...
    Resource Type:
    Dissertation
    Full Text:
    from training conditions, G, D, and D++ performance degrades rapidly. Using CKI, agents are provided

  • Robust Long Term Autonomy Through Local Behaviors

    Creator:
    Yates, Connor L.
    Abstract:
    Multi-robot teams offer promising solutions for many long term deployments in remote and dangerous domains, such as extraterrestrial or underseas exploration. However, long term deployments present many problems preventing robot teams from operating effectively. Learning over long time scales is makes it difficult to assign credit to robots' actions, as...
    Resource Type:
    Dissertation
    Full Text:
    this specific order, starting at the first node (G = 0). All parent nodes must be observed before each

  • Providing Informative Feedback for Learning in Tightly Coupled Multiagent Domains

    Creator:
    Rockefeller, Golden C.
    Abstract:
    Autonomous agents that sense, decide, act, and coordinate effectively with each other are critical in many real-world domains such as autonomous driving, search and rescue missions, air traffic management, and underwater or deep space exploration. All such domains share a key difficulty: though high-level mission goals are clear to system...
    Resource Type:
    Dissertation

  • Multi-Reward Learning and Sparse Rewards

    Creator:
    Christopher, Reid M.
    Abstract:
    Reinforcement learning has made impressive strides in solving problems in challenging domains, but problems are increasingly being described with sparse rewards. Sparse rewards directly reduce the rate at which useful feedback is provided to the learner and make it difficult to distinguish between what specific actions led to the reception...
    Resource Type:
    Capstone Project

  • Tackling Credit Assignment Using Memory and Multilevel Optimization for Multiagent Reinforcement Learning

    Creator:
    Khadka, Shauharda
    Abstract:
    There is growing commercial interest in the use of multiagent systems in real world applications. Some examples include inventory management in warehouses, smart homes, planetary exploration, search and rescue, air-traffic management and autonomous transportation systems. However, multiagent coordination is an extremely challenging problem. First, information relevant for coordination is often...
    Resource Type:
    Dissertation

  • Learning in Complex Domains : Leveraging Multiple Rewards through Alignment

    Creator:
    Klinkhammer, Eric R.
    Abstract:
    In this thesis, we introduce alignment-based algorithms for improving the performance of reinforcement learning solutions for problems where the reward signal cannot be collapsed into a single number. Many real world problems require an agent to balance performance, longevity, and safety, and do so across different timelines. The key to...
    Resource Type:
    Masters Thesis
    Full Text:
    gradient of G strongly match. . . . . . . . 51 LIST OF FIGURES (Continued) Figure Page 6.6 The

  • Distortion of Agent States for Improved Coordination

    Creator:
    Yates, Connor L
    Abstract:
    Many real world problems have partial solutions or intermediary steps which can lead toward solving the problem. When assembling robotic teams to solve these problems, we have intuition about which intermediary steps are more useful than others. We examine methods to identify and apply our designer intuition onto tightly coupled...
    Resource Type:
    Honors College Thesis
    Full Text:
    quickly that what it is doing is correct. The global reward (G), and the difference reward (D) can be

  • D₊₊ : Structural Credit Assignment in Tightly Coupled Multiagent Domains

    Creator:
    Rahmattalabi, Aida
    Abstract:
    Autonomous multiagent teams can be used in complex exploration tasks to both expedite the exploration and improve the efficiency. However, use of multiagent systems presents additional challenges. Specifically, in domains where the agents' actions are tightly coupled, coordinating multiple agents to achieve cooperative behavior at the group level is difficult....
    Resource Type:
    Masters Thesis
    Full Text:
    observations to be made to account as success. . . . 25 4.2 Observation performance of policies trained on G

  • Multiagent learning for locomotion and coordination in tensegrity robotics

    Creator:
    Iscen, Atil
    Abstract:
    Tensegrity structures are composed of pure compressional elements that are connected via a network of pure tensional elements. The concept of tensegrity promises numerous advantages to the field of robotics. Tensegrity robots are, however, notoriously difficult to control due to their oscillatory nature and nonlinear interaction between the components. Multiagent...
    Resource Type:
    Dissertation
    Full Text:
    different corners. D, G and L represent Difference, Global and Local Rewards re- spectively. The results