Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 12 of 12 for “"multiarmed"”.

  1. NON-STATIONARY MULTIARMED BANDITS FOR SATIATION AND SEASONALITY PHENOMENA IN MUSIC RECOMMENDER SYSTEMS

    … into theoretical aspects of non-stationary multiarmed bandits, motivated by their application to music recommender systems. An intrinsic challenge of such systems lies in evolving user preferences. Rather than finding a single optimal item, the objective is to craft an ordered sequence of …

    milano Repository record for NON-STATIONARY MULTIARMED BANDITS FOR SATIATION AND SEASONALITY PHENOMENA IN MUSIC RECOMMENDER SYSTEMS (opens in a new tab)

  2. Algoritmos para multi-armed bandits: teoria e aplicação à precificação dinâmica

    … decision-making, focusing specifically on the multiarmed bandit (MAB) framework. In its classical formulation, the MAB problem involves an agent facing a row of slot machines (bandits), with a limited number of pulls (arms) available. The agent’s goal is to determine a sequence of actions that …

    brazil-uerj Repository record for Algoritmos para multi-armed bandits: teoria e aplicação à precificação dinâmica (opens in a new tab)

  3. Thresholded linear bandits

    … lies in the intersection of several important multiarmed bandit (MAB) variants, including active learning, structured bandits, and learning halfspaces. To achieve sublinear regret in the presence of exponentially many arms, one method is to exploit the structure of the reward function. However, …

    uvic Repository record for Thresholded linear bandits (opens in a new tab)

  4. Planning for an adaptive evader with application to drug interdiction operations

    … Force assets. This model is akin to the multiarmed bandit problem. We create two algorithms for the Adapting Evader Model. One algorithm uses an optimal learning policy and the other uses a heuristic learning policy. We also create an algorithm for the interdiction model using the …

    nps Repository record for Planning for an adaptive evader with application to drug interdiction operations (opens in a new tab)

  5. Multipath QUIC Implementation and Scheduling with Adversarial Multi-Armed Bandits

    … scheduler, Multipath Scheduler with Adversarial MultiArmed Bandit (MSAB), is presented, which makes an intelligent access network selection and is conscious of network heterogeneity and dynamics. Extensive simulation results with ns-3 show that the proposed MSAB scheduler can outperform …

    uvic Repository record for Multipath QUIC Implementation and Scheduling with Adversarial Multi-Armed Bandits (opens in a new tab)

  6. PEO and PEO-heparin modified surfaces for blood contacting applications

    … weights, and different end groups, to star-type, multiarmed PEOs, to linear PEOs in which one end bears a thiol group and the other a bioactive group. Different reaction protocols were used for the synthesis of the different PEOs. Terminal thiol groups were attached to PEO by reaction with …

    mcmaster Repository record for PEO and PEO-heparin modified surfaces for blood contacting applications (opens in a new tab)

  7. Analytics for Sustainable and Retail Operations

    … budget. We formulate the problem as a restless multiarmed bandit model and develop an index-based inspection policy. We also demonstrate how to extend the model to incorporate heterogeneous inspection costs and the possibility of environmental disaster. Simulations using data from EPA Region 2 …

    gatech Repository record for Analytics for Sustainable and Retail Operations (opens in a new tab)

  8. Learning In The Presence Of Unawareness

    … problems with unawareness (BLPUs), which are multiarmed bandit problems in which there may be arms that the DM is unaware of. We provide a policy that is 0.25c-optimal for BLPUs, where c is a constant determined by the probability of discovering a new arm and the probability of there being …

    cornell Repository record for Learning In The Presence Of Unawareness (opens in a new tab)

  9. Multi-Armed Bandits with Risk-Aware Performance Measures

    The classical multi-armed bandit problem involves a learner and a collection of arms with unknown reward distributions. At each round, the learner selects an arm and receives new information. The learner faces a tradeoff between exploiting the current information and exploring all arms. The …

    waterloo Repository record for Multi-Armed Bandits with Risk-Aware Performance Measures (opens in a new tab)

  10. Decentralized multi-user multi-armed bandits with user dependent reward distributions

    The uncoordinated spectrum access problem is studied using a multi-player multi-armed bandits framework. We consider a decentralized multi-player stochastic multi-armed bandit model where the players cannot communicate with each other and can observe only their own actions and rewards. Furthermore, …

    uiuc Repository record for Decentralized multi-user multi-armed bandits with user dependent reward distributions (opens in a new tab)

  11. Off-policy latent variable modeling for fast bandit personalization

    Medical treatments are decided based on medical history and the current symptoms of a patient. For chronic illnesses this can be difficult, as long-time patients develop an amount of medical data that is hard to grasp. We propose the use of machine-learning methods to both condense this …

    chalmers Repository record for Off-policy latent variable modeling for fast bandit personalization (opens in a new tab)

  12. Scaling Up Reinforcement Learning without Sacrificing Optimality by Constraining Exploration

    The purpose of this dissertation is to understand how algorithms can efficiently learn to solve new tasks based on previous experience, instead of being explicitly programmed with a solution for each task that we want it to solve. Here a task is a series of decisions, such as a robot vacuum …

    tamu Repository record for Scaling Up Reinforcement Learning without Sacrificing Optimality by Constraining Exploration (opens in a new tab)